Skip to content

Conversation

@AstroBoy1
Copy link

The gradients for the loss on the weights is not needed during validation and test because no weight updates are made during those steps. I added the no_grad context manager: https://pytorch.org/docs/stable/generated/torch.no_grad.html

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.

1 participant