[hotfix] Quantization patch; fix semantic_search_faiss/semantic_search_usearch rescoring
#2558
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
tomaarsen
added a commit
that referenced
this pull request
Mar 26, 2024
…rch_usearch rescoring (#2558) * Correctly update the scores after rescoring * Sort scores from high to low in recommended example
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Outdated suggestions cannot be applied.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.
Hello!
Pull Request overview
semantic_search_recommended.pyDetails
Fix semantic_search_faiss/semantic_search_usearch rescoring
In these helper methods, the rescoring would update the corpus indices based on the new scores, but the returned scores corresponded to the original ranking, not the new one. As a result, the returned score for the first sample would not strictly have to be the highest, even if it should be.
This has been resolved now.
Fix incorrect sorting in
semantic_search_recommended.pyThe rescoring results were sorted the wrong way around: low to high rather than high to low. This resulted in worse performance than should be expected. This is also resolved with this PR.