This repository was archived by the owner on Jan 2, 2025. It is now read-only.
Hybrid search (lexical + semantic) merged with RRF #1154
Merged
Conversation
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
rm some unwanted files rm some unwanted files
|
|
oppiliappan
reviewed
Nov 29, 2023
oppiliappan
approved these changes
Nov 30, 2023
Comment on lines
+557
to
+559
| .iter() | ||
| .take(limit.try_into().unwrap()) | ||
| .cloned() |
Contributor
There was a problem hiding this comment.
can consume here instead of cloning?
Suggested change
| .iter() | |
| .take(limit.try_into().unwrap()) | |
| .cloned() | |
| .into_iter() | |
| .take(limit.try_into().unwrap()) |
ggordonhall
approved these changes
Dec 1, 2023
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
3 participants
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Outdated suggestions cannot be applied.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.
Hybrid search replaces semantic search with a mix between semantic search results and lexical search results Results are merged using reciprocal rank fusion (RRF).
To be able to merge results, each search (semantic and lexical) have to operate on the same documents. Here the documents are code snippets and to makes things easier, we use Qdrant as vector and snippet DB/index manager.
Wdrant lexical search does not rank lexical results based on known algorithms (tf-idf, bm25). So we rerank lexical results using word counts and number of query unique words matched.
This PR: