[gpt2pre 1.2] Start End Packer call method #7791
Merged
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
mattsoulanille
requested changes
Jul 8, 2023
|
@mattsoulanille your comments have been addressed. Thanks for the review! |
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Outdated suggestions cannot be applied.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.
Implements the full StartEndPacker class including it's
call()method.This is PR 2/2 of the full StartEndPacker implementation.
Depends on #7790.
NOTE:
The keras implementation includes a
return_padding_maskparameter in the constructor. Setting thisTruemakes the call method return the padded ragged tensor AND a mask of padded tokens as a tuple. This tuple return was not possible in the TypeScript implementation without changing thecalltype signature to[Tensor|Tensor[], Tensor|Tensor[]]. Therefore, I opted to create a separate method,callAndReturnPaddingMask()that returns just that. Thecallmethod simply takes the first result of this method call.Room for optimization:
Currently, a mask will be calculated every time
callis called, even when not needed. Decomposing the function further can help and can be done in the cleanup PR.