Skip to content

Conversation

parhamfh
Copy link

At least in the case of the autoencoder demo callback it will consume batches from the same underlying data queue as the train/val DataLoader, which prevents them from being used in training/validation.

This proposes to create a distinct DataLoader for the demo callback.

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