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

Delay device mesh import #561

Merged
merged 4 commits into from
Apr 26, 2024
Merged

Delay device mesh import #561

merged 4 commits into from
Apr 26, 2024

Conversation

2015aroras
Copy link
Collaborator

DeviceMesh is introduced in torch 2.2, and so trying to import it in torch 2.1 is caused the training code to break. This PR delays its import until it is needed (for hybrid sharding).

Fixes #559

@2015aroras 2015aroras marked this pull request as ready for review April 26, 2024 16:15
@2015aroras 2015aroras merged commit 295d309 into main Apr 26, 2024
9 of 10 checks passed
@2015aroras 2015aroras deleted the shanea/delay-device-mesh-import branch April 26, 2024 19:51
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.

No module named 'torch.distributed.device_mesh'
1 participant