[Dynamo] Fix TIMM benchmark compute_loss#97423
Closed
yanboliang wants to merge 3 commits intopytorch:masterfrom
Closed
[Dynamo] Fix TIMM benchmark compute_loss#97423yanboliang wants to merge 3 commits intopytorch:masterfrom
yanboliang wants to merge 3 commits intopytorch:masterfrom
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/97423
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit da505cf: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Chillee
approved these changes
Mar 23, 2023
Collaborator
Chillee
left a comment
There was a problem hiding this comment.
:O
Perhaps I just got unlucky 🤔 I was looking at mnasnet, and I noticed that both the loss and the output are quite a bit less accurate.
Contributor
Author
|
@pytorchbot merge |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fixes #97382
#95416 fixed a critical bug in dynamo benchmark, where AMP tests fall back to eager mode before that PR. However, after that PR, we found a list of TIMM models amp + eager + training testing failed.
Now we identified the root cause is: high loss values make gradient checking harder, as small changes in accumulation order upset accuracy checks. We should switch to the helper function
reduce_to_scalar_losswhich has been used by Torchbench tests.After switching to
reduce_to_scalar_loss, TIMM models accuracy pass rate grows from 67.74% to 91.94% in my local test. The rest 5 failed models(ese_vovnet19b_dw, fbnetc_100, mnasnet_100, mobilevit_s, sebotnet33ts_256) need further investigation and handling, but I think it should be similar reason.cc @soumith @voznesenskym @penguinwu @anijain2305 @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @Xia-Weiwen @wenzhe-nrv @jiayisunx @desertfire