Fix FlaxDataCollatorForT5MLM for texts with "<pad>" #15443
Closed
Conversation
This file contains hidden or 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
If the input text contains "<pad>" (true for C4/en dataset) T5 tokenizer assigns it index 0. Collator code (line 371) assumes that only tokens with `id > 0` are the text tokens and the rest are sentinel ids (`-1`), which turns out not to be true for this particular case. This causes the line to throw an error similar to ``` input_ids = input_ids_full[input_ids_full >= 0].reshape((batch_size, -1)) *** ValueError: cannot reshape array of size 1040383 into shape (8192,newaxis) ``` This modification should also allow to use padding in T5 pre-training.
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. |
Contributor
|
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread. Please note that issues that do not follow the contributing guidelines are likely to be ignored. |
Author
|
@patil-suraj , cold you please review this PR? It's very short |
Collaborator
|
Oh we have a similar PR here from @patrickvonplaten |
Author
|
Great! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.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.
If the input text contains
"<pad>"(true for C4/en dataset) T5 tokenizer assigns it index 0.Collator code (line 371) assumes that only tokens with
id > 0are the text tokens and the rest are sentinel ids (-1), which turns out not to be true for this particular case.This causes the line to throw an error similar to
This modification should also allow to use padding in T5 pre-training.
@patil-suraj (note edited by @stas00 to tag the correct maintainer for flax)