[fx] Add use_reentrant=False to checkpoint in codegen#1463
Merged
FrankLeeeee merged 23 commits intohpcaitech:mainfrom Aug 17, 2022
Merged
[fx] Add use_reentrant=False to checkpoint in codegen#1463FrankLeeeee merged 23 commits intohpcaitech:mainfrom
FrankLeeeee merged 23 commits intohpcaitech:mainfrom
Conversation
Merge ColossalAI
Daily merge
FrankLeeeee
approved these changes
Aug 17, 2022
Contributor
|
I suggest you should test on torchvision models instead of customized models. |
Contributor
Author
I think we could add some compatibility test in the future, I think the customized model could be short and efficiently covers the situation we might encounter |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The previous PR #1460 added
use_reentrant=Falsefeature in checkpoint, to address the problems we encountered in activation searching strategy in #1451. In this PR, I bring this feature to our ColoTracer codegen process. Once the activation_codegen detects that one checkpoint's input will be used in an in-place operation, it will useuse_reentrant=Falseoption when calling thecolossalai.utils.activation_checkpoint.checkpoint, otherwise it will useuse_reentrant=Truewhich consists with the original checkpoint function.