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.

Sign up for GitHub

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

Jump to bottom

[FSDP][optim_state_dict] Make FSDP optim_state_dict aware of DDP prefix #96415

Closed

fegin wants to merge 1 commit into pytorch:master from fegin:export-D43893609

Contributor

fegin commented Mar 9, 2023

Summary: When wrapping FSDP within DDP, optimizer state_dict may be broken due to the prefix of DDP. This PR fixes the issue.

Test Plan: CI

Differential Revision: D43893609

fegin requested review from mrshenli, zhaojuanmao, rohan-varma, H-Huang, awgu, kwen2501 and wanchaol as code owners

March 9, 2023 12:11

pytorch-bot bot commented Mar 9, 2023 •

edited

Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/96415

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 0a2ac3e:
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

pytorch-bot bot added the release notes: distributed (fsdp) label

facebook-github-bot added the fb-exported label

Contributor

facebook-github-bot commented Mar 9, 2023

This pull request was exported from Phabricator. Differential Revision: D43893609

zhaojuanmao approved these changes

View reviewed changes

fegin added the ciflow/trunk label

Contributor

facebook-github-bot commented Mar 9, 2023

This pull request was exported from Phabricator. Differential Revision: D43893609

fegin force-pushed the export-D43893609 branch from 2dfaa35 to d321430 Compare

March 9, 2023 18:28

fegin added a commit to fegin/pytorch that referenced this pull request


          [FSDP][optim_state_dict] Make FSDP optim_state_dict aware of DDP pref…

348c7c2

…ix (pytorch#96415)

Summary:
Pull Request resolved: pytorch#96415

When wrapping FSDP within DDP, optimizer state_dict may be broken due to the prefix of DDP. This PR fixes the issue.

Test Plan: CI

Reviewed By: zhaojuanmao

Differential Revision: D43893609

fbshipit-source-id: 32039229f56c052797b558e0266a5798332d792e

fegin force-pushed the export-D43893609 branch from d321430 to 348c7c2 Compare

March 9, 2023 18:40

Contributor

facebook-github-bot commented Mar 9, 2023

This pull request was exported from Phabricator. Differential Revision: D43893609

1 similar comment

Contributor

facebook-github-bot commented Mar 10, 2023

This pull request was exported from Phabricator. Differential Revision: D43893609

fegin force-pushed the export-D43893609 branch from 348c7c2 to ba53283 Compare

March 10, 2023 18:27


          [FSDP][optim_state_dict] Make FSDP optim_state_dict aware of DDP pref…

0a2ac3e

…ix (pytorch#96415)

Summary:
Pull Request resolved: pytorch#96415

When wrapping FSDP within DDP, optimizer state_dict may be broken due to the prefix of DDP. This PR fixes the issue.

Test Plan: CI

Reviewed By: zhaojuanmao

Differential Revision: D43893609

fbshipit-source-id: efac71301a919bf57f63f3f2bfd4795d0ebb3e83

Contributor

facebook-github-bot commented Mar 10, 2023

This pull request was exported from Phabricator. Differential Revision: D43893609

fegin force-pushed the export-D43893609 branch from ba53283 to 0a2ac3e Compare

March 10, 2023 18:36

Contributor

facebook-github-bot commented Mar 13, 2023

@pytorchbot merge

(Initiating merge automatically since Phabricator Diff has merged)

Collaborator

pytorchmergebot commented Mar 13, 2023

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

pytorchmergebot added the Merged label

pytorchmergebot closed this in

f5a0b31

cyyever pushed a commit to cyyever/pytorch_private that referenced this pull request


          [FSDP][optim_state_dict] Make FSDP optim_state_dict aware of DDP pref…

1705f80

…ix (#96415)

Summary: When wrapping FSDP within DDP, optimizer state_dict may be broken due to the prefix of DDP. This PR fixes the issue.

Test Plan: CI

Differential Revision: D43893609

Pull Request resolved: pytorch/pytorch#96415
Approved by: https://github.com/zhaojuanmao

cyyever pushed a commit to cyyever/pytorch_private that referenced this pull request


          [FSDP][optim_state_dict] Make FSDP optim_state_dict aware of DDP pref…

921de41

…ix (#96415)

Summary: When wrapping FSDP within DDP, optimizer state_dict may be broken due to the prefix of DDP. This PR fixes the issue.

Test Plan: CI

Differential Revision: D43893609

Pull Request resolved: pytorch/pytorch#96415
Approved by: https://github.com/zhaojuanmao

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

zhaojuanmao zhaojuanmao approved these changes

mrshenli Awaiting requested review from mrshenli

rohan-varma Awaiting requested review from rohan-varma

H-Huang Awaiting requested review from H-Huang

awgu Awaiting requested review from awgu

kwen2501 Awaiting requested review from kwen2501

wanchaol Awaiting requested review from wanchaol

Labels

ciflow/trunk fb-exported Merged release notes: distributed (fsdp)