[Distributed] refine training_pipeline_step to avoid memory leaks #11107
Conversation
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
tianshuo78520a
merged commit cb431d3
into
PaddlePaddle:incubate/paddlenlp-fleety_20250421
6 of 7 checks passed
|
Thanks for your contribution! |
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Outdated suggestions cannot be applied.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.
PR types
Function optimization
PR changes
Others
Description
Pass a local function to forward_backward_pipeline instead of the dataset itself. This prevents the dataset from being passed as a direct argument to forward_backward_pipeline, which would create additional reference counts that cannot be cleared, leading to GPU memory leaks.
This PR depends on PaddlePaddle/Paddle#75446.
This PR is cp from #11103