Revert the default reranker to NaiveReranker.#277
Merged
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #277 +/- ##
=======================================
Coverage 99.66% 99.66%
=======================================
Files 24 24
Lines 1788 1788
=======================================
Hits 1782 1782
Misses 6 6 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
It's been observed that the
NaiveRerankeroutperforms the defaultCrossEncoderRerankermodel in a lot of scenarios. While the cross encoders might have more potential, it doesn't provide a good OOTB experience (and changing them won't break anything). Thus, I'm reverting the default reranker toNaiveReranker, which is faster, less computationally intensive and more consistent for 0-config usage.