Adding more filter tasks and filter recall for so_vector#789
Merged
benwtrent merged 4 commits intoelastic:masterfrom Jun 12, 2025
Merged
Adding more filter tasks and filter recall for so_vector#789benwtrent merged 4 commits intoelastic:masterfrom
benwtrent merged 4 commits intoelastic:masterfrom
Conversation
so_vector/operations/default.json
Outdated
Contributor
There was a problem hiding this comment.
nit: formatting on some of these looks wonky
Member
Author
There was a problem hiding this comment.
I ran the formatter, I am surprised it didn't remove tabs and replace with spaces :)
mayya-sharipova
approved these changes
Jun 11, 2025
Contributor
mayya-sharipova
left a comment
mayya-sharipova
left a comment
There was a problem hiding this comment.
@benwtrent Thanks, nice enhancement of the track!
Indeed, would be nice to cache exact neighbours for filtered queries, as they need to be executed twice.
| "k": 10, | ||
| "num_candidates": 50, | ||
| "filter": { | ||
| "range": { |
Contributor
There was a problem hiding this comment.
Nit: looks like formatting should be improved
| } | ||
| } | ||
| }, | ||
| { |
Contributor
There was a problem hiding this comment.
Should we add "include-in-reporting": false to all recall operations>
tteofili
approved these changes
Jun 12, 2025
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
4 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Here I am adding some more filter searches (knn only) and recall tasks for each of them.
I think 10% and 20% hit difficult spots for HNSW and are important to calculate. making them random ensures we don't hit "nice" pockets of the graph where all the nearest neighbors actually match the filter criteria already or a particular part of the graph actually matches filter criteria.