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

Reading small datasets forces reshuffle and extra epochs despite number of epochs set #10

Open
XapaJIaMnu opened this issue Mar 22, 2023 · 0 comments
Labels
enhancement New feature or request minor

Comments

@XapaJIaMnu
Copy link
Contributor

When testing with very short datasets (10 lines), the trainer wraps the dataset 10 times before it stops reading so that it can fill in a full batch.

Unfortunately this won't be as easy as the dataset itself has no notion of the maximum number of epochs it's supposed to read so it can't throw StopIteration exception

@XapaJIaMnu XapaJIaMnu added enhancement New feature or request minor labels Mar 22, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request minor
Projects
None yet
Development

No branches or pull requests

1 participant