-
Notifications
You must be signed in to change notification settings - Fork 4
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Add jit macro for simplified testing (#194)
- Loading branch information
Showing
2 changed files
with
26 additions
and
2 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
6866f05
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)
1331366172
ns1332810111
ns1.00
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant
1333147663
ns1310849226
ns1.02
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme)
1530087148
ns1398288716
ns1.09
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme)
3054505913
ns2623910053
ns1.16
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Lux
230499086
ns215139523
ns1.07
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme)
5310260095
ns5334887020
ns1.00
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant
5117023688
ns5125148459
ns1.00
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme)
5696115262
ns5128099715
ns1.11
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme)
6877852849
ns7093041527
ns0.97
ViT base (256 x 256 x 3 x 32)/forward/CPU/Lux
31332890298
ns31469979010
ns1.00
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme)
1377593493
ns1390179366
ns0.99
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant
1419824350
ns1268184459.5
ns1.12
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme)
1370284838
ns1270873858.5
ns1.08
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme)
2678832914
ns2492061660
ns1.07
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Lux
8674244
ns8221824
ns1.06
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme)
1711793645
ns1711092077
ns1.00
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant
1593142398
ns1558493025
ns1.02
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme)
1553322361
ns1546123882
ns1.00
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme)
2788658689
ns2735606253
ns1.02
ViT small (256 x 256 x 3 x 4)/forward/CPU/Lux
3829077050
ns2456759080
ns1.56
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :after_enzyme)
1292126503
ns1286461508.5
ns1.00
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant
1255035565.5
ns1287819878
ns0.97
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme)
1241708897.5
ns1229738497.5
ns1.01
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme)
2627736445
ns2423291169
ns1.08
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Lux
21072763
ns20850848
ns1.01
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme)
2141201304
ns2148735048
ns1.00
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant
2150005669
ns2138885245
ns1.01
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme)
2153988118
ns2133624257
ns1.01
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme)
3403989575
ns3388591813
ns1.00
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Lux
6112741293.5
ns5994160924
ns1.02
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme)
1319271524
ns1312161644.5
ns1.01
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant
1321146725.5
ns1286524449.5
ns1.03
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme)
1336696942
ns1304833938.5
ns1.02
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme)
2956277164
ns2654597387
ns1.11
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Lux
7206908
ns7062374
ns1.02
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme)
1446752820
ns1463335858
ns0.99
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant
1420199736
ns1418279069
ns1.00
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme)
1419361542
ns1407215120
ns1.01
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme)
2620658225
ns2610977913
ns1.00
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Lux
1343552629
ns1313118431
ns1.02
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme)
1339740837
ns1266033162
ns1.06
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant
1297025776.5
ns1246060690
ns1.04
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme)
1307783663.5
ns1334724021
ns0.98
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme)
2422904622
ns2615371467
ns0.93
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Lux
13705203.5
ns11338191
ns1.21
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme)
1689997595
ns1712640221
ns0.99
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant
1709524846
ns1714535441
ns1.00
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme)
1699909787
ns1699386874
ns1.00
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme)
2910898852
ns2934490244
ns0.99
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Lux
3172820959
ns3109868396.5
ns1.02
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme)
1272957214
ns1304791782
ns0.98
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant
1288787604
ns1285502833
ns1.00
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme)
1292898756
ns1277688710
ns1.01
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme)
2528780375
ns2599975980
ns0.97
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Lux
25554565
ns25551082.5
ns1.00
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme)
2158163346
ns2164823256
ns1.00
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant
2164587691
ns2168156538
ns1.00
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme)
2157132364
ns2195060353
ns0.98
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme)
3395183827
ns3415103088
ns0.99
ViT small (256 x 256 x 3 x 16)/forward/CPU/Lux
6334070118
ns6792188737
ns0.93
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :after_enzyme)
1228186625
ns1251711662
ns0.98
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant
1202678487.5
ns1314443020
ns0.91
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme)
1189947014.5
ns1296451333
ns0.92
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme)
2346429480
ns2569030485
ns0.91
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Lux
50144272.5
ns50146964
ns1.00
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme)
2973474372
ns3044550425
ns0.98
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant
3005386081
ns3049319408
ns0.99
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme)
2959359145
ns2999963120
ns0.99
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme)
4361243109
ns4363286803
ns1.00
ViT small (256 x 256 x 3 x 32)/forward/CPU/Lux
9378921820
ns10042481459
ns0.93
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme)
1247107066
ns1302500995
ns0.96
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant
1226562970
ns1297504625
ns0.95
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme)
1232975234.5
ns1310827906
ns0.94
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme)
2344405060
ns2446072271
ns0.96
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Lux
67888329.5
ns67921126
ns1.00
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme)
3168087012
ns3156309083
ns1.00
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant
3147204975
ns3173315537
ns0.99
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme)
3238059309
ns3130494484
ns1.03
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme)
4525581424
ns4504318538
ns1.00
ViT base (256 x 256 x 3 x 16)/forward/CPU/Lux
13793978482
ns14749507671
ns0.94
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme)
1191983955
ns1303044095
ns0.91
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant
1202390873
ns1304317104.5
ns0.92
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme)
1204541731
ns1332711162
ns0.90
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme)
2466522108
ns2642760989
ns0.93
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Lux
19336923
ns19441467
ns0.99
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme)
2207876303
ns1863044822
ns1.19
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant
1998857639
ns1837806913
ns1.09
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme)
1904292223
ns1855662911
ns1.03
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme)
3061997168
ns3030096911
ns1.01
ViT base (256 x 256 x 3 x 4)/forward/CPU/Lux
3696727697
ns3331794574
ns1.11
This comment was automatically generated by workflow using github-action-benchmark.