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.

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

duplicate models when using pytorch lightning with ddp accelerator #292

Closed
tankeco opened this issue Jan 21, 2021 · 3 comments
Closed

duplicate models when using pytorch lightning with ddp accelerator #292

tankeco opened this issue Jan 21, 2021 · 3 comments

Comments

@tankeco
Copy link
Contributor

tankeco commented Jan 21, 2021 •
edited
Loading

When using pytorch lightning with ddp accelerator on 4 gpus, I find every checkpoint is recorded 4 times on web UI with different ids. One is on default_output_uri (s3://...) and three are on file:///... .

4

Note pytorch lightning automatically ensures that the model is saved only on the main process: https://pytorch-lightning.readthedocs.io/en/stable/weights_loading.html#manual-saving-with-accelerators
clearml-bot pushed a commit that referenced this issue Jan 24, 2021
Fix issue #292 pytorch-lighting multi node store
381ae21
@bmartinn
Copy link
Member

Hi @tankeco
Could you maybe test the fix? it's already part of the latest RC

pip install clearml==0.17.5rc2

@tankeco
Copy link
Contributor Author

tankeco commented Jan 25, 2021

it fixed :)

@bmartinn
Copy link
Member

Great news @tankeco :) closing this issue.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@tankeco @bmartinn