Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Check status of broken tests #742

Merged
merged 3 commits into from
Jul 1, 2024
Merged

Check status of broken tests #742

merged 3 commits into from
Jul 1, 2024

Conversation

avik-pal
Copy link
Member

@avik-pal avik-pal commented Jun 30, 2024

  • AMDGPU is working for almost all cases now. Conv doesn't work great because MIOpen only supports CrossCor. We rewrite Conv to CrossCor for AMDGPU usecases.
  • Distributed -- Still causes deadlocks?
  • All other broken tests (except some of the loss tests) have been fixed.

Copy link

codecov bot commented Jun 30, 2024

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.82%. Comparing base (1a61165) to head (4839c75).

Current head 4839c75 differs from pull request most recent head 054ecfa

Please upload reports for the commit 054ecfa to get more accurate results.

Additional details and impacted files
@@            Coverage Diff             @@
##             main     #742      +/-   ##
==========================================
- Coverage   96.40%   95.82%   -0.58%     
==========================================
  Files          54       54              
  Lines        2726     2731       +5     
==========================================
- Hits         2628     2617      -11     
- Misses         98      114      +16     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

test/utils_tests.jl Outdated Show resolved Hide resolved
test/layers/conv_tests.jl Outdated Show resolved Hide resolved
test/layers/conv_tests.jl Outdated Show resolved Hide resolved
test/layers/conv_tests.jl Outdated Show resolved Hide resolved
test/layers/conv_tests.jl Outdated Show resolved Hide resolved
@avik-pal avik-pal force-pushed the ap/amd_tests branch 4 times, most recently from 1189ae3 to f6ad1ae Compare June 30, 2024 22:18
Copy link
Contributor

@github-actions github-actions bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Benchmark Results

Benchmark suite Current: cf8cc9b Previous: 1a61165 Ratio
Dense(2 => 2)/cpu/reverse/ReverseDiff (compiled)/(2, 128) 3524.4444444444443 ns 3658.125 ns 0.96
Dense(2 => 2)/cpu/reverse/Zygote/(2, 128) 6971.714285714285 ns 7185.4 ns 0.97
Dense(2 => 2)/cpu/reverse/Tracker/(2, 128) 20689 ns 21610 ns 0.96
Dense(2 => 2)/cpu/reverse/ReverseDiff/(2, 128) 9347.75 ns 9954.4 ns 0.94
Dense(2 => 2)/cpu/reverse/Flux/(2, 128) 8710.4 ns 8986.625 ns 0.97
Dense(2 => 2)/cpu/reverse/SimpleChains/(2, 128) 4361.555555555556 ns 4523.375 ns 0.96
Dense(2 => 2)/cpu/reverse/Enzyme/(2, 128) 1144.5975609756097 ns 1162.9044117647059 ns 0.98
Dense(2 => 2)/cpu/forward/NamedTuple/(2, 128) 1165.2767295597484 ns 1112.9741935483871 ns 1.05
Dense(2 => 2)/cpu/forward/ComponentArray/(2, 128) 1164.0243902439024 ns 1163.123287671233 ns 1.00
Dense(2 => 2)/cpu/forward/Flux/(2, 128) 1773.3936170212767 ns 1774.3508771929824 ns 1.00
Dense(2 => 2)/cpu/forward/SimpleChains/(2, 128) 179.75647668393782 ns 180.07943262411348 ns 1.00
Dense(20 => 20)/cpu/reverse/ReverseDiff (compiled)/(20, 128) 17283 ns 17313 ns 1.00
Dense(20 => 20)/cpu/reverse/Zygote/(20, 128) 16902 ns 16952 ns 1.00
Dense(20 => 20)/cpu/reverse/Tracker/(20, 128) 35712.5 ns 39563 ns 0.90
Dense(20 => 20)/cpu/reverse/ReverseDiff/(20, 128) 27961.5 ns 29234 ns 0.96
Dense(20 => 20)/cpu/reverse/Flux/(20, 128) 20067 ns 21470 ns 0.93
Dense(20 => 20)/cpu/reverse/SimpleChains/(20, 128) 16872 ns 17242 ns 0.98
Dense(20 => 20)/cpu/reverse/Enzyme/(20, 128) 4276.75 ns 4325.285714285715 ns 0.99
Dense(20 => 20)/cpu/forward/NamedTuple/(20, 128) 3794.625 ns 3880.875 ns 0.98
Dense(20 => 20)/cpu/forward/ComponentArray/(20, 128) 3849.75 ns 3936 ns 0.98
Dense(20 => 20)/cpu/forward/Flux/(20, 128) 4651.571428571428 ns 4889 ns 0.95
Dense(20 => 20)/cpu/forward/SimpleChains/(20, 128) 1648.1 ns 1663.1 ns 0.99
Conv((3, 3), 3 => 3)/cpu/reverse/ReverseDiff (compiled)/(64, 64, 3, 128) 38197991 ns 41122113 ns 0.93
Conv((3, 3), 3 => 3)/cpu/reverse/Zygote/(64, 64, 3, 128) 56410980.5 ns 57716330 ns 0.98
Conv((3, 3), 3 => 3)/cpu/reverse/Tracker/(64, 64, 3, 128) 74996972 ns 81840420 ns 0.92
Conv((3, 3), 3 => 3)/cpu/reverse/ReverseDiff/(64, 64, 3, 128) 87968256 ns 91437945.5 ns 0.96
Conv((3, 3), 3 => 3)/cpu/reverse/Flux/(64, 64, 3, 128) 71032153.5 ns 75926448 ns 0.94
Conv((3, 3), 3 => 3)/cpu/reverse/SimpleChains/(64, 64, 3, 128) 11742738 ns 11662872.5 ns 1.01
Conv((3, 3), 3 => 3)/cpu/reverse/Enzyme/(64, 64, 3, 128) 17689740 ns 17787447 ns 0.99
Conv((3, 3), 3 => 3)/cpu/forward/NamedTuple/(64, 64, 3, 128) 6917850 ns 7032740 ns 0.98
Conv((3, 3), 3 => 3)/cpu/forward/ComponentArray/(64, 64, 3, 128) 6864588.5 ns 6996461 ns 0.98
Conv((3, 3), 3 => 3)/cpu/forward/Flux/(64, 64, 3, 128) 9930672 ns 12114935 ns 0.82
Conv((3, 3), 3 => 3)/cpu/forward/SimpleChains/(64, 64, 3, 128) 6313050.5 ns 6387272 ns 0.99
vgg16/cpu/reverse/Zygote/(32, 32, 3, 16) 734571226 ns 767844159 ns 0.96
vgg16/cpu/reverse/Zygote/(32, 32, 3, 64) 2501522227 ns 2602811621 ns 0.96
vgg16/cpu/reverse/Zygote/(32, 32, 3, 2) 128457533.5 ns 144091616 ns 0.89
vgg16/cpu/reverse/Tracker/(32, 32, 3, 16) 851885210 ns 893102323 ns 0.95
vgg16/cpu/reverse/Tracker/(32, 32, 3, 64) 2948844279 ns 3246592469 ns 0.91
vgg16/cpu/reverse/Tracker/(32, 32, 3, 2) 205481768.5 ns 222301050.5 ns 0.92
vgg16/cpu/reverse/Flux/(32, 32, 3, 16) 675097423.5 ns 723583721 ns 0.93
vgg16/cpu/reverse/Flux/(32, 32, 3, 64) 2623332472.5 ns 2632526936 ns 1.00
vgg16/cpu/reverse/Flux/(32, 32, 3, 2) 124332100 ns 126310974.5 ns 0.98
vgg16/cpu/forward/NamedTuple/(32, 32, 3, 16) 170157527 ns 176488869 ns 0.96
vgg16/cpu/forward/NamedTuple/(32, 32, 3, 64) 637526461.5 ns 649092634 ns 0.98
vgg16/cpu/forward/NamedTuple/(32, 32, 3, 2) 44286540.5 ns 34416017 ns 1.29
vgg16/cpu/forward/ComponentArray/(32, 32, 3, 16) 163802469.5 ns 163630557 ns 1.00
vgg16/cpu/forward/ComponentArray/(32, 32, 3, 64) 627314805 ns 640318006 ns 0.98
vgg16/cpu/forward/ComponentArray/(32, 32, 3, 2) 29615145 ns 29797077.5 ns 0.99
vgg16/cpu/forward/Flux/(32, 32, 3, 16) 184654418 ns 225565619 ns 0.82
vgg16/cpu/forward/Flux/(32, 32, 3, 64) 700606803.5 ns 851688514 ns 0.82
vgg16/cpu/forward/Flux/(32, 32, 3, 2) 36815776 ns 38079734.5 ns 0.97
Conv((3, 3), 64 => 64)/cpu/reverse/ReverseDiff (compiled)/(64, 64, 64, 128) 1275335870 ns 1261029479 ns 1.01
Conv((3, 3), 64 => 64)/cpu/reverse/Zygote/(64, 64, 64, 128) 1815208670 ns 1854091416 ns 0.98
Conv((3, 3), 64 => 64)/cpu/reverse/Tracker/(64, 64, 64, 128) 2388461028 ns 2553348643 ns 0.94
Conv((3, 3), 64 => 64)/cpu/reverse/ReverseDiff/(64, 64, 64, 128) 2347678421 ns 2528691188 ns 0.93
Conv((3, 3), 64 => 64)/cpu/reverse/Flux/(64, 64, 64, 128) 1859431510.5 ns 1925582416 ns 0.97
Conv((3, 3), 64 => 64)/cpu/reverse/Enzyme/(64, 64, 64, 128) 547695707 ns 562461809 ns 0.97
Conv((3, 3), 64 => 64)/cpu/forward/NamedTuple/(64, 64, 64, 128) 312358337 ns 320567710 ns 0.97
Conv((3, 3), 64 => 64)/cpu/forward/ComponentArray/(64, 64, 64, 128) 313655398 ns 318656712 ns 0.98
Conv((3, 3), 64 => 64)/cpu/forward/Flux/(64, 64, 64, 128) 343033906.5 ns 379455021 ns 0.90
Conv((3, 3), 1 => 1)/cpu/reverse/ReverseDiff (compiled)/(64, 64, 1, 128) 11630854.5 ns 12014151 ns 0.97
Conv((3, 3), 1 => 1)/cpu/reverse/Zygote/(64, 64, 1, 128) 17586854 ns 17777207 ns 0.99
Conv((3, 3), 1 => 1)/cpu/reverse/Tracker/(64, 64, 1, 128) 18791974 ns 18982505.5 ns 0.99
Conv((3, 3), 1 => 1)/cpu/reverse/ReverseDiff/(64, 64, 1, 128) 23611215 ns 23814355 ns 0.99
Conv((3, 3), 1 => 1)/cpu/reverse/Flux/(64, 64, 1, 128) 17667905.5 ns 17776145 ns 0.99
Conv((3, 3), 1 => 1)/cpu/reverse/SimpleChains/(64, 64, 1, 128) 1146731 ns 1163571 ns 0.99
Conv((3, 3), 1 => 1)/cpu/reverse/Enzyme/(64, 64, 1, 128) 5666409 ns 5835756.5 ns 0.97
Conv((3, 3), 1 => 1)/cpu/forward/NamedTuple/(64, 64, 1, 128) 2034012.5 ns 2049324.5 ns 0.99
Conv((3, 3), 1 => 1)/cpu/forward/ComponentArray/(64, 64, 1, 128) 2020371 ns 2030939 ns 0.99
Conv((3, 3), 1 => 1)/cpu/forward/Flux/(64, 64, 1, 128) 2063949.5 ns 2074821 ns 0.99
Conv((3, 3), 1 => 1)/cpu/forward/SimpleChains/(64, 64, 1, 128) 196390 ns 202707 ns 0.97
Dense(200 => 200)/cpu/reverse/ReverseDiff (compiled)/(200, 128) 292230 ns 291222 ns 1.00
Dense(200 => 200)/cpu/reverse/Zygote/(200, 128) 263446 ns 265629.5 ns 0.99
Dense(200 => 200)/cpu/reverse/Tracker/(200, 128) 365337.5 ns 364529 ns 1.00
Dense(200 => 200)/cpu/reverse/ReverseDiff/(200, 128) 406405 ns 407123 ns 1.00
Dense(200 => 200)/cpu/reverse/Flux/(200, 128) 270569 ns 273649.5 ns 0.99
Dense(200 => 200)/cpu/reverse/SimpleChains/(200, 128) 401947 ns 407198 ns 0.99
Dense(200 => 200)/cpu/reverse/Enzyme/(200, 128) 83056 ns 83466 ns 1.00
Dense(200 => 200)/cpu/forward/NamedTuple/(200, 128) 80341 ns 81592 ns 0.98
Dense(200 => 200)/cpu/forward/ComponentArray/(200, 128) 80612 ns 82003 ns 0.98
Dense(200 => 200)/cpu/forward/Flux/(200, 128) 86242 ns 86571 ns 1.00
Dense(200 => 200)/cpu/forward/SimpleChains/(200, 128) 104356 ns 104455 ns 1.00
Conv((3, 3), 16 => 16)/cpu/reverse/ReverseDiff (compiled)/(64, 64, 16, 128) 181218320 ns 200999670 ns 0.90
Conv((3, 3), 16 => 16)/cpu/reverse/Zygote/(64, 64, 16, 128) 318828611 ns 324420843 ns 0.98
Conv((3, 3), 16 => 16)/cpu/reverse/Tracker/(64, 64, 16, 128) 368589427 ns 417216292 ns 0.88
Conv((3, 3), 16 => 16)/cpu/reverse/ReverseDiff/(64, 64, 16, 128) 445881116 ns 459212467 ns 0.97
Conv((3, 3), 16 => 16)/cpu/reverse/Flux/(64, 64, 16, 128) 368652441 ns 391131363 ns 0.94
Conv((3, 3), 16 => 16)/cpu/reverse/SimpleChains/(64, 64, 16, 128) 304658232 ns 320280603 ns 0.95
Conv((3, 3), 16 => 16)/cpu/reverse/Enzyme/(64, 64, 16, 128) 100248184 ns 102161858 ns 0.98
Conv((3, 3), 16 => 16)/cpu/forward/NamedTuple/(64, 64, 16, 128) 43264955 ns 43990307.5 ns 0.98
Conv((3, 3), 16 => 16)/cpu/forward/ComponentArray/(64, 64, 16, 128) 42951815 ns 43757610.5 ns 0.98
Conv((3, 3), 16 => 16)/cpu/forward/Flux/(64, 64, 16, 128) 49468908 ns 57183117.5 ns 0.87
Conv((3, 3), 16 => 16)/cpu/forward/SimpleChains/(64, 64, 16, 128) 27960494 ns 28324912 ns 0.99
Dense(2000 => 2000)/cpu/reverse/ReverseDiff (compiled)/(2000, 128) 20146432 ns 18822434 ns 1.07
Dense(2000 => 2000)/cpu/reverse/Zygote/(2000, 128) 19300959 ns 19540731 ns 0.99
Dense(2000 => 2000)/cpu/reverse/Tracker/(2000, 128) 23095286.5 ns 23356974 ns 0.99
Dense(2000 => 2000)/cpu/reverse/ReverseDiff/(2000, 128) 23980499 ns 24094714 ns 1.00
Dense(2000 => 2000)/cpu/reverse/Flux/(2000, 128) 19329285 ns 19581858.5 ns 0.99
Dense(2000 => 2000)/cpu/reverse/Enzyme/(2000, 128) 6432044 ns 6520727 ns 0.99
Dense(2000 => 2000)/cpu/forward/NamedTuple/(2000, 128) 6434439 ns 6499891.5 ns 0.99
Dense(2000 => 2000)/cpu/forward/ComponentArray/(2000, 128) 6396926 ns 6481714 ns 0.99
Dense(2000 => 2000)/cpu/forward/Flux/(2000, 128) 6462751 ns 6495020 ns 1.00

This comment was automatically generated by workflow using github-action-benchmark.

@avik-pal avik-pal force-pushed the ap/amd_tests branch 6 times, most recently from 6d12cbf to 4839c75 Compare July 1, 2024 00:34
@avik-pal avik-pal force-pushed the ap/amd_tests branch 2 times, most recently from 054ecfa to cf8cc9b Compare July 1, 2024 01:09
@avik-pal avik-pal merged commit 4505c9f into main Jul 1, 2024
46 of 57 checks passed
@avik-pal avik-pal deleted the ap/amd_tests branch July 1, 2024 01:10
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant