You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
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
The text was updated successfully, but these errors were encountered:
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
The text was updated successfully, but these errors were encountered: