Bugfix: last minibatch padding #8
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ed this from final logits/labels
|
|
AlexTMallen
reviewed
Feb 12, 2024
| # Ensure only the actual predictions and labels are extended, | ||
| # not the padded ones | ||
| all_logits.extend(logits[:mbatch_len].to(io_device)) | ||
| all_labels.extend(labels[:mbatch_len]) | ||
| all_logits = torch.stack(all_logits) |
There was a problem hiding this comment.
One edge case to consider is what happens if mbatch_len is 0
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
To avoid the error
TypeError: TransformerWithHead.forward() missing 1 required positional argument: 'input_ids', similar to pytorch/pytorch#15161 when using dataparallel and the last minibatch gets truncated, we add right-padding to input_ids and then take off the extra logits/labels that this creates.