Skip to content

Added a check in no_sync() to avoid errors when using deepspeed zero2/3 - #3656

Merged
S1ro1 merged 1 commit into
huggingface:mainfrom
xliu0105:main
Jun 26, 2025
Merged

S1ro1 merged 1 commit into
huggingface:mainfrom
xliu0105:main

Conversation

@xliu0105

@xliu0105 xliu0105 commented Jun 26, 2025 •

Copy link
Copy Markdown
Contributor

Added a check in no_sync() to avoid errors when using deepspeed zero2/3

What does this PR do?

When using accelerate and deepspeed with zero2 or zero3, and setting the gradient accumulation > 1, an error will be reported: AssertionError: no_sync context manager is incompatible with gradient partitioning logic of ZeRO stage 2

Therefore, modify the no_sync() function and add a judgment to it to ensure that no_sync does not conflict with the implementation of zero2 and zero3.

Thanks @alisafaya

Fixes #3481

Before submitting

  • This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
  • Did you read the contributor guideline,
    Pull Request section?
  • Was this discussed/approved via a Github issue or the forum? Please add a link
    to it if that's the case.
  • Did you make sure to update the documentation with your changes? Here are the
    documentation guidelines, and
    here are tips on formatting docstrings.
  • Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.
@SunMarc @zach-huggingface

@S1ro1 S1ro1 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you!

@HuggingFaceDocBuilderDev

Copy link
Copy Markdown

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

@S1ro1
S1ro1 merged commit 175fe91 into huggingface:main Jun 26, 2025
kashif pushed a commit to kashif/accelerate that referenced this pull request Jul 6, 2025
S1ro1 pushed a commit that referenced this pull request Jul 9, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

DeepSpeed ZeRO Stage 2/3 incompatibility with no_sync context manager

3 participants