Sparse idf dot #4126
Merged
Sparse idf dot #4126
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
…mix them up on the type level
xzfc
requested changes
Apr 29, 2024
agourlay
reviewed
Apr 29, 2024
| @@ -254,10 +307,9 @@ impl SegmentsSearcher { | |||
| search_in_segment( | |||
| segment, | |||
| partial_batch_request, | |||
| 0, | |||
There was a problem hiding this comment.
We used to pass explicitly zero in that case vs available_point_count for the query above.
Now both search_in_segment will use the same value from the query_context.
Not sure of the consequences yet.
There was a problem hiding this comment.
None, if the heuristic is not used, it won't use this parameter
agourlay
reviewed
Apr 29, 2024
Co-authored-by: Arnaud Gourlay <arnaud.gourlay@gmail.com>
|
FYI the sparse vector congruence tests are still green with this branch 👌 |
agourlay
approved these changes
Apr 29, 2024
xzfc
approved these changes
Apr 29, 2024
generall
added a commit
that referenced
this pull request
May 2, 2024
* introduce QueryContext, which accumulates runtime info needed for executing search * fmt * propagate query context into segment internals * [WIP] prepare idf stats for search query context * Split SparseVector and RemmapedSparseVector to guarantee we will not mix them up on the type level * implement filling of the query context with IDF statistics * implement re-weighting of the sparse query with idf * fmt * update idf param only if explicitly specified (more consistent with diff param update * replace idf bool with modifier enum, improve further extensibility * test and fixes * Update lib/collection/src/operations/types.rs Co-authored-by: Arnaud Gourlay <arnaud.gourlay@gmail.com> * review fixes * fmt --------- Co-authored-by: Arnaud Gourlay <arnaud.gourlay@gmail.com>
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.
This PR introduces a
modifierparam, which can be configured for individual sparse vectors.It is also introduces a single non-empty value for the modifier, called
idf.Usage example:
Modifier alters the weights of sparse vectors, so the resulting score (and order) are different.
The
idfmodifier multiplies weights of the sparse vectors to account for "rarity" of the vector elements in the current collection. Less frequent the element is, more impact on the scoring it will have.Exact formula looks like this:
where
n- number of points in shard,df- number of points with given element.This component is computed at query time and multiplied with query before performing the actual search.
Similar scoring component is used in Okapi BM25, which allows usage of sparse vectors as a replacement for Bm25 text search without the necessity of pre-computing of the corpus statistics.