Update the sync batch norm implementation for sync_bn#66
Closed
renganxu wants to merge 1 commit intofacebookresearch:mainfrom
Closed
Update the sync batch norm implementation for sync_bn#66renganxu wants to merge 1 commit intofacebookresearch:mainfrom
renganxu wants to merge 1 commit intofacebookresearch:mainfrom
Conversation
Summary: The sync_bn batch norm has been using a non-efficient implementation NaiveSyncBatchNorm. This is because in pytorch<=1.5, nn.SyncBatchNorm had incorrect gradient when the batch size on each worker is different, so a NaiveSyncBatchNorm was implemented. This issue has been fixed in pytorch/pytorch#36382. So we change that implementation back to the faster nn.SyncBatchNorm. Since nn.SyncBatchNorm only has GPU implementation, for the unit tests running with single process on CPU, the original NaiveSyncBatchNorm is used. Reviewed By: wat3rBro Differential Revision: D35300913 fbshipit-source-id: c649f4dfc1ab5b2a0dd91d0187b072ae2de229ff
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D35300913 |
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary:
The sync_bn batch norm has been using a non-efficient implementation NaiveSyncBatchNorm. This is because in pytorch<=1.5, nn.SyncBatchNorm had incorrect gradient when the batch size on each worker is different, so a NaiveSyncBatchNorm was implemented. This issue has been fixed in pytorch/pytorch#36382. So we change that implementation back to the faster nn.SyncBatchNorm.
Since nn.SyncBatchNorm only has GPU implementation, for the unit tests running with single process on CPU, the original NaiveSyncBatchNorm is used.
Reviewed By: wat3rBro
Differential Revision: D35300913