Open
Conversation
- Implemented for model loading to prevent redundant IO and parsing when creating multiple instances. - Added auto-download capability to : now automatically downloads/copies remote files (e.g., gs://) to the local cache if missing. - Refactored to separate model loading logic into standalone cached functions. - Updated to verify download behavior.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 : This PR updates the ParquetDataSource class in: gemma/gm/data/_parquet.py to correctly handle a list of file paths. Previously, passing a list of paths would cause the initialization/reading to fail or behave unexpectedly as it was not iterating through the paths. The new implementation iterates through provided paths, reads each table individually, and concatenates them using pyarrow.concat_tables.
Changes:
Verification:
Verified manually using a reproduction script (mocked) demonstrating that read_table is now called for each path in the list and concat_tables is utilized.