Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix FSDP auto_wrap using characters instead of full str for layers #3075

Merged
merged 2 commits into from
Sep 4, 2024

Conversation

muellerzr
Copy link
Collaborator

What does this PR do?

Recent release broke FSDP for models like llama that define a set of layers to split by. It'd parse the string as a tuple of characters instead. This PR fixes it + adjusts the tests to be more strict (tests bert and llama for auto wrap policy)

Fixes critical issue of FSDP being broken on the release.

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

@HuggingFaceDocBuilderDev

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.

Copy link
Member

@SunMarc SunMarc left a comment

Choose a reason for hiding this comment

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

LGTM ! Thanks for fixing !

@muellerzr muellerzr merged commit ab89fc7 into main Sep 4, 2024
28 checks passed
@muellerzr muellerzr deleted the patchfix-fsdp branch September 4, 2024 16:44
@anhuong
Copy link

anhuong commented Sep 5, 2024

Thank you so much for this fix! When can we expect to have this change in a release?

@muellerzr
Copy link
Collaborator Author

@anhuong we're waiting on one other PR to pass its tests, then I'll be doing a patchfix today

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.

4 participants