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

Enable loading full optimizer checkpoints with FSDP #17747

Merged
merged 6 commits into from
Jun 10, 2023

Conversation

leng-yue
Copy link
Contributor

@leng-yue leng-yue commented Jun 3, 2023

What does this PR do?

Making restoring optimizer states from full checkpoint possible, which is a supplement of #17623.

Before submitting
  • Was this discussed/agreed via a GitHub issue? (not for typos and docs)
  • Did you read the contributor guideline, Pull Request section?
  • Did you make sure your PR does only one thing, instead of bundling different changes together?
  • Did you make sure to update the documentation with your changes? (if necessary)
  • Did you write any new necessary tests? (not for typos and docs)
  • Did you verify new and existing tests pass locally with your changes?
  • Did you list all the breaking changes introduced by this pull request?
  • Did you update the CHANGELOG? (not for typos, docs, test updates, or minor internal changes/refactors)

PR review

Anyone in the community is welcome to review the PR.
Before you start reviewing, make sure you have read the review guidelines. In short, see the following bullet-list:

Reviewer checklist
  • Is this pull request ready for review? (if not, please submit in draft mode)
  • Check that all items from Before submitting are resolved
  • Make sure the title is self-explanatory and the description concisely explains the PR
  • Add labels and milestones (and optionally projects) to the PR so it can be classified

@github-actions github-actions bot added the fabric lightning.fabric.Fabric label Jun 3, 2023
@leng-yue leng-yue marked this pull request as ready for review June 3, 2023 09:16
@awaelchli awaelchli added strategy: fsdp Fully Sharded Data Parallel community This PR is from the community labels Jun 6, 2023
@awaelchli awaelchli added this to the 2.1 milestone Jun 6, 2023
@awaelchli awaelchli self-assigned this Jun 6, 2023
@mergify mergify bot added the has conflicts label Jun 6, 2023
@leng-yue
Copy link
Contributor Author

leng-yue commented Jun 7, 2023

Any update?

@mergify mergify bot removed the has conflicts label Jun 7, 2023
Copy link
Contributor

@awaelchli awaelchli left a comment

Choose a reason for hiding this comment

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

@leng-yue That's awesome, thank you so much for jumping on this so quickly! I have minor comments. Great work and thanks for extending the test!

src/lightning/fabric/strategies/fsdp.py Outdated Show resolved Hide resolved
tests/tests_fabric/strategies/test_fsdp_integration.py Outdated Show resolved Hide resolved
Copy link
Contributor

@awaelchli awaelchli left a comment

Choose a reason for hiding this comment

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

Amazing!

@Borda Borda enabled auto-merge (squash) June 9, 2023 07:09
@mergify mergify bot added the ready PRs ready to be merged label Jun 9, 2023
src/lightning/fabric/strategies/fsdp.py Outdated Show resolved Hide resolved
src/lightning/fabric/strategies/fsdp.py Outdated Show resolved Hide resolved
src/lightning/fabric/strategies/fsdp.py Outdated Show resolved Hide resolved
src/lightning/fabric/strategies/fsdp.py Outdated Show resolved Hide resolved
auto-merge was automatically disabled June 9, 2023 17:25

Head branch was pushed to by a user without write access

@leng-yue
Copy link
Contributor Author

I don't know why some tasks failed after I changed few lines, and there is no error message in azure pipelines...

@awaelchli awaelchli enabled auto-merge (squash) June 10, 2023 11:12
@awaelchli awaelchli merged commit a23bae3 into Lightning-AI:master Jun 10, 2023
ishandutta0098 pushed a commit to ishandutta0098/lightning that referenced this pull request Jun 10, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
community This PR is from the community fabric lightning.fabric.Fabric ready PRs ready to be merged strategy: fsdp Fully Sharded Data Parallel
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants