Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Update MLIR bindings regeneration workflow to include package update …
…and precompile step
…and precompile step
e084f9e
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Reactant.jl Benchmarks
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :after_enzyme)
1350735972
ns1351952654
ns1.00
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant
1204673780
ns1329740254
ns0.91
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme)
1222340041
ns1313154586
ns0.93
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme)
2330048189
ns2566992420
ns0.91
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Lux
219309808
ns221738431
ns0.99
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme)
5215948763
ns5795183922
ns0.90
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant
6254101434
ns5365250430
ns1.17
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme)
5201734622
ns5580309503
ns0.93
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme)
7460701472
ns7018981334
ns1.06
ViT base (256 x 256 x 3 x 32)/forward/CPU/Lux
30735680811
ns32330341377
ns0.95
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme)
1332885325
ns1295393361
ns1.03
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant
1298553637
ns1391655239.5
ns0.93
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme)
1340798546.5
ns1288112883.5
ns1.04
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme)
2364390446
ns2580913506
ns0.92
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Lux
8812264.5
ns8622025
ns1.02
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme)
1635773422
ns1638985456
ns1.00
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant
1621494476
ns1598870688
ns1.01
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme)
1609772124
ns1604887372
ns1.00
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme)
2830061683
ns2845843527
ns0.99
ViT small (256 x 256 x 3 x 4)/forward/CPU/Lux
2877521675
ns3290222049
ns0.87
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :after_enzyme)
1310197829
ns1337995509
ns0.98
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant
1304840979.5
ns1266589678.5
ns1.03
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme)
1238123602.5
ns1304533265
ns0.95
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme)
2594487217
ns2619821981
ns0.99
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Lux
22665498
ns22673367
ns1.00
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme)
2163350638
ns2164741585
ns1.00
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant
2227062239
ns2187101004
ns1.02
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme)
2171553343
ns2222453725
ns0.98
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme)
3558585268
ns3473087162
ns1.02
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Lux
7747467151.5
ns5922666949.5
ns1.31
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme)
1283318913
ns1409824378
ns0.91
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant
1216591656.5
ns1274413695
ns0.95
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme)
1370999738
ns1229630167.5
ns1.11
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme)
2428504820
ns2435178422
ns1.00
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Lux
7186419
ns6720828
ns1.07
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme)
1467788340
ns1443071529
ns1.02
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant
1449340906
ns1455478624
ns1.00
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme)
1443567043
ns1436315353
ns1.01
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme)
2702131547
ns2698020568
ns1.00
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Lux
1095865115
ns1179865891
ns0.93
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme)
1210758212
ns1311903715.5
ns0.92
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant
1233393897
ns1208899239.5
ns1.02
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme)
1225142106.5
ns1315408397
ns0.93
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme)
2374047383
ns2622407972
ns0.91
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Lux
12131784
ns12273085
ns0.99
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme)
1757223123
ns1747750553
ns1.01
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant
1753678153
ns1759657061
ns1.00
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme)
1747793752
ns1730465781
ns1.01
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme)
3015378976
ns2979617791
ns1.01
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Lux
2910278832
ns2985205457
ns0.97
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme)
1322233467
ns1205936693
ns1.10
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant
1322254383
ns1289931712
ns1.03
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme)
1199964457
ns1274603974.5
ns0.94
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme)
2400347071
ns2353800786
ns1.02
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Lux
27163900
ns27361038
ns0.99
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme)
2214089958
ns2245983275
ns0.99
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant
2239028395
ns2235719581
ns1.00
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme)
2212157343
ns2225764215
ns0.99
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme)
3527800382
ns3488656246
ns1.01
ViT small (256 x 256 x 3 x 16)/forward/CPU/Lux
5811115543.5
ns6519742812.5
ns0.89
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :after_enzyme)
1226440738
ns1273072927
ns0.96
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant
1160529232.5
ns1275903559
ns0.91
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme)
1284160996.5
ns1306679492
ns0.98
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme)
2249108448
ns2418323014
ns0.93
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Lux
52687663
ns52729682
ns1.00
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme)
3019468880
ns3106843960
ns0.97
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant
3069044359
ns3110765368
ns0.99
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme)
3044926123
ns3067530015
ns0.99
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme)
4458047333
ns4512690716
ns0.99
ViT small (256 x 256 x 3 x 32)/forward/CPU/Lux
8346217965
ns23082341445
ns0.36
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme)
1248064916
ns1244033717
ns1.00
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant
1265067809.5
ns1220583584
ns1.04
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme)
1274798656.5
ns1241043964.5
ns1.03
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme)
2435701548
ns2534565534
ns0.96
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Lux
70937902
ns71029594
ns1.00
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme)
3268589369
ns3308971321
ns0.99
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant
3233353228
ns3272699344
ns0.99
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme)
3264816245
ns3281133548
ns1.00
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme)
4685900819
ns4663116034
ns1.00
ViT base (256 x 256 x 3 x 16)/forward/CPU/Lux
15059348114
ns13237204037
ns1.14
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme)
1293711217
ns1206105557
ns1.07
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant
1320275711
ns1205282113
ns1.10
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme)
1275143769
ns1254093260
ns1.02
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme)
2411331594
ns2518219292
ns0.96
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Lux
20583569
ns20789324
ns0.99
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme)
2119165721
ns1913409296
ns1.11
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant
2218102344
ns1887591901
ns1.18
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme)
2186413098
ns1917404653
ns1.14
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme)
3188249362
ns3181159514
ns1.00
ViT base (256 x 256 x 3 x 4)/forward/CPU/Lux
3741883940
ns3522016222.5
ns1.06
This comment was automatically generated by workflow using github-action-benchmark.