0.7.4

Hello, AutoML folk! We’re releasing a minor version of FEDOT that includes the following.

PyPi release: https://pypi.org/project/fedot/0.7.4/

Features & Enhancements:

• Boosting method implementation (XGBoost) by @RomanKharkovskoy in #1209
• Boosting method implementation (LightGBM) by @RomanKharkovskoy in #1264

Bugfixes:

• Sktime fix by @nicl-nno in #1323

What's Changed

• fix: NaNs containing forecast on some ets configs by @Lopa10ko in #1292
• fix: NaNs containing data on fit stage for fast_ica by @Lopa10ko in #1295
• fix: go for forecast method instead of predict in ets by @Lopa10ko in #1299
• fix: indexing and seasonal_periods check that endog is at least twice the size by @Lopa10ko in #1305
• Fix multitarget stratification by @DRMPN in #1308
• Update unit-build.yml by @technocreep in #1294
• Hotfix of fit time estimation for initial assumption by @nicl-nno in #1309
• Fix for #1296 by @aPovidlo in #1312
• hotfix: go for FeaturesType instead of InputData in a pipeline tuning by @Lopa10ko in #1311
• Boosting method implementation (XGBoost) by @RomanKharkovskoy in #1209
• fix: bring back cgru integration tests by @Lopa10ko in #1313
• Boosting method implementation (LightGBM) by @RomanKharkovskoy in #1264
• hotfix: Bug with target encoding by @aPovidlo in #1314
• hotfix: boostings related failure of test_*_models_check_fit_predict_correct by @Lopa10ko in #1316
• hotfix: add punkt_tab to nltk LookupError check by @Lopa10ko in #1319
• hotfix: pin nltk version to avoid downloading punkt_tab by @Lopa10ko in #1321
• Sktime fix by @nicl-nno in #1323
• m4 timeseries benchmark by @Lopa10ko in #1277
• fix: do not launch composition for atomized model by @DRMPN in #1324

Full Changelog: v0.7.3.2...v0.7.4

0.7.3.2

Minor fixes and improvements.

What's Changed

• Update main_concepts.rst by @glad4enkonm in #1275
• Fix pca by @valer1435 in #1267
• Fix sporadic forecasting test failure by @Lopa10ko in #1280
• Update data.py by @v1docq in #1283
• Add validation on _window_size param in topo extractor by @Lopa10ko in #1284
• Update version of codecov action by @DRMPN in #1289
• Fix gitlab mirroring by @Nunkyl in #1288
• Fix gensim for PyPi release by @DRMPN in #1287

Full Changelog: v0.7.3.1...v0.7.3.2

0.7.3.1

Minor fixes and improvements, better installation

0.7.3

0.7.3 Latest

Hello, AutoML folk! We’re releasing a minor version of FEDOT that includes the following.

PyPi release: https://pypi.org/project/fedot/0.7.3/

Features & Enhancements:

• Update of GOLEM (core framework) dependency to 0.4.0 version, that has some important features itself
• Improvement of data preprocessing
• Improvements for time series forecasting
• Better API

Full list of changes:

• Fix params setting by @YamLyubov in #1121
• Clean up SA module & add example by @maypink in #1114
• Debug mode and modified fitness evaluation for for time series by @nicl-nno in #1116
• Lagged implementation speedup by @kasyanovse in #1139
• Support of multimodal data in DataSourceSplitter + bugfix #1119 by @kasyanovse in #1138
• Fix low speed of lagged implementation by @kasyanovse in #1144
• 628 bugfix by @kasyanovse in #1145
• Fix random FEDOT results with set seed by @kasyanovse in #1143
• Replace exogenous example by @YamLyubov in #1108
• Fix multiobj tuning by @YamLyubov in #1156
• Decimal places control implemented by @nicl-nno in #1161
• Title for Pipeline plots by @maypink in #1162
• Prediction Intervals by @GrigoriJasnovidov in #1149
• Add MASE metric by @valer1435 in #1166
• Remove redundant metrics mapping by @YamLyubov in #1167
• Refactor convert to op by @valer1435 in #1176
• Fix default tuning params by @valer1435 in #1175
• Add desc to not implemented errors by @valer1435 in #1177
• from fedot import Fedot by @MorrisNein in #1179
• Boosting method implementation (Catboost) by @aPovidlo in #1155
• Add documented API params setters by @MorrisNein in #1077
• add InputData from pd and numpy by @ChrisLisbon in #1184
• Support for changes in GOLEM by @kasyanovse in #1189
• Docs fix for atomized model by @nicl-nno in #1197
• get predict_data.num_classes if trained_model havent num_classes by @ElishaFlacon in #1199
• Metrics fixes, tests, refactoring, typing improvement by @MorrisNein in #1214
• 922 preprocessor acceleration by @IIaKyJIuH in #1004
• Add direct examples by @valer1435 in #1192
• Improve the rule for time series pipeline verification by @kasyanovse in #1210
• Fixes for unit and intergration tests by @nicl-nno in #1238
• add window size selector by @kasyanovse in #1237
• Add topofeatures by @valer1435 in #1241

Full Changelog: v0.7.2...v0.7.3

0.7.2

Hello, AutoML folk! We’re releasing a minor version of FEDOT that includes the following.

PyPi release: https://pypi.org/project/fedot/0.7.2/

Features & Enhancements:

• Update of GOLEM (core framework) dependency to 0.3.2 version, that has some important features itself

Bugfixes:

• Minor improvements and fixes

0.7.1

Hello, AutoML folk! We’re releasing a minor version of FEDOT that includes the following.

PyPi release: https://pypi.org/project/fedot/0.7.1/

Features & Enhancements:

• Update of GOLEM (core framework) dependency to 0.3.1 version, that has some important features itself
• Added Meta Rules that automatically select best parameters based on dataset: with rules for early_stopping_generations, preset and cross-validation folds were added (#1057)
• Improved API parameters and made documentation more clear and structured (#1067, #1041)
• Improve test suite and its performance (#1098)
• Improved DataMerger for textual data – now multi-column text table can be used with FEDOT (#1052)

Bugfixes:

• Bug with wrong combinations of operations in pipelines for time series forecasting was fixed.
• Multiple initial assumptions support was fixed (#1070)
• Various minor fixes

GOLEM core 0.7.0

Hi, folk!

This release marks major change! Our team separated all the core modules (graph, adapter, optimizer, tuner etc.) into the separate project GOLEM (https://github.com/aimclub/GOLEM).

FEDOT now contains modules related to Data handling, preprocessing, machine learning logic like Pipeline (implementation of ML Graph), ML operations, ML metrics.

There're also few other changes:

• #1029 -- CLSTM model refactoring
• #1046 -- Transition of GitHub tests from Python 3.7 to 3.8

Self-sufficient 0.6.2

Hi all!

Importantly, release 0.6.2 marks the last self-sufficient version of FEDOT before transition to GOLEM optimization core (https://github.com/aimclub/GOLEM).

This release introduces a number of API enhancements and several bug fixes.
Enhancements:

• #1017 -- Now it's possible to enable memory analytic for debugging the framework performance.
• #1019 -- It became easier and more effective to work with large datasets.
• #1021 -- PipelineBuilder now can be used for any Graphs with a help of appropriate Adapter.
• #990 -- A number of user API improvements.
• #1025 -- Data preprocessing is now optional and can be disabled through new API parameter.
• #1031 -- Refactoring of Pipeline Node classes, that now are much simplified.

Bug fixes:

• #1010 -- Tuning is more benefitial now with correct metric deviation computation.
• #1012
• #1022
• #1023

Version 0.6.1

Hi, folk!
We're making a new minor release with a number of improvements. This is an important release in a sense that this is a last release of self-contained FEDOT. The next major release will mark a separation of the optimizer core into the separate project.

New features, better quality & changes in API

• More intuitive predict interface for time series forecasting (#930)
• Pipeline save/load now have more intuitive behavior (#971)
• Early stopping criteria now can take timeout into considerations, and not only number of iterations (early_stopping_timeout api parameter)
• Graph nodes now can be accessed by name or uid (#982)
• Tuner speed is better due to better initial params in the search space (#985)

Enhancements and fixes:

• Fix inplace modification of data during data definition (resolves #943)
• Fix regression preprocessing (#955)
• Less evaluation errors during population selection in corner cases (#956)
• Fix getting suitable operations for multi ts (#981)
• Integration tests are fixed & passing now
• More minor fixes & minor class interface refactorings
• Important fix for multi-objective optimization (#996)

Documentation is extended

Architectural refactorings are continued:

• Better PipelineAdapter (#941)
• Abstracting optimiser core (most tasks in issue #713 are done)
  Notably, Serializer subsystem is now extendable (#969)

v0.6.0

Hi everyone!
We released a new major version of FEDOT - 0.6.0

It includes a lot of major changes:

• Improvement of API for multi-modal datasets and models;
• New PipelineBuilder (#597) – that simplifies manual construction of ML Pipelines;
• Joblib was embedded as a multiprocessing backend (#843). Data exchange between processes minimized (#926);
• Embedding stratify k fold strategy for cases with imbalance data;
• New visualization of graphs, pipelines and optimisation history;

Also, this release contains by a lot of architectural refactorings of the framework:

• New Graph Adapter subsystem (#876);
• Merging two different implementation of evolutionary optimizer (parameter-free & usual) into one EvoGraphOptimizer (#687)
• Architectural refactorings of the Graph hierarchy (#750)
• Introduce notions of Objective & Fitness (#654) – classes that substitutes simple float metric values & abstract single vs. multi-objective metrics
• Refactored parameter classes – for more intuitive segregation of different parameters controlling optimization process (#852)
• Refactored DataMerger facility
• Refactoring of selection operator implementation (#918)

Also, there are various bug-fixes related to ML operations, evolutionary operators & internal Graph operations.