Skip to content

Conversation

JWHennessey
Copy link

Currently when resuming from a checkpoint too much GPU memory is used. This work around resolves the issue.

@rosinality
Copy link
Owner

Ah, thank you! I think you can resolve this using torch.load(args.ckpt, map_location=lambda storage, loc: storage).

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.

2 participants