[scorer] - fix: padding tokens - #83
Merged
Merged
Conversation
…token indices - Implement utility function to get a fallback pad token ID when necessary, ensuring compatibility with various tokenizers - Refine the replacement of `IGNORE_INDEX` with appropriate token IDs across different methods to address special cases and simplify the logic - Replace hardcoded ignore index checks and token ID replacements with function calls to enhance code maintainability and readability
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Improves handling of special token indices in LMScorer by introducing utility functions
_get_fallback_pad_token_id()and_replace_ignore_index(). Replaces hardcoded-100(ignore index) checks with reusable helper functions, adds fallback logic to handle tokenizers without apad_token_id(tryingeos_token_id,unk_token_id, or defaulting to 0), and improves type hints. Enhances robustness when decoding labels that contain ignore indices.Risk
Low - refactoring existing logic into helper functions with improved fallback handling. The behavior remains the same for tokenizers with
pad_token_iddefined, and adds graceful fallback for edge cases.Tests
Existing LMScorer tests cover the functionality.