ORTModule memory efficient gradient management #18907
Closed
Conversation
This file contains 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
pengwa
added
the
training
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
orttraining/orttraining/python/training/ortmodule/_mem_efficent_training.py
Fixed
Show fixed
Hide fixed
orttraining/orttraining/python/training/ortmodule/_mem_efficent_training.py
Fixed
Show fixed
Hide fixed
pengwa
changed the title
pengwa
force-pushed
the
pengwa/mem_efficient_grad_mgr
branch
from
c585bc8 to
be122e3
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
ORTModule memory efficient gradient management
Previously I have tried to solve the coarsed-grained gradient accumulation/update problem in ORTModule with #8979, while that resolution somehow is not fully validated with DDP or there is user hooks on the gradient accumulation on torch parameter.
This PR is addressing the problem in the similar approach as PR 8979, e.g. trigger gradient accumulation once ORT computed the grad, but instead of use a
AccumulateGradop, this time with a ONNX operatorPythonOp, internally it will callparam.backward(grad), which will help handle all related hooks correctly.Replacedby #18924
Motivation and Context