Repository navigation
Conversation
ts_euclidean_distance_SIMD stops once its partial sum reaches the bound and returns that partial sum. The top-k scans accepted results with `dist <= knn[k-1]`, so a series whose abandoned partial sum equalled the current bound was inserted with that under-estimated distance, and its position was then reported next to a distance that belongs to another series. With the position-based de-duplication this wrong entry could also block the real series, because it already occupied the same position slot. Compare with `<` in the approximate scan and in the MESSI L2 leaf scan, so a distance that is not strictly better than the current k-th is never inserted. On 200k clustered 128-d vectors, k=10: returned (id, distance) pairs whose distance does not match the series: 8 of 50000 before, 0 after; fraction of the true top-10 returned: 0.9998 -> 1.0000. On 1M vectors the same: 1.0000. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-authored-by: sophisid <55842445+sophisid@users.noreply.github.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
ts_euclidean_distance_SIMD stops once its partial sum reaches the bound and returns that partial sum. The top-k scans accepted results with
dist <= knn[k-1], so a series whose abandoned partial sum equalled the current bound was inserted with that under-estimated distance, and its position was then reported next to a distance that belongs to another series. With the position-based de-duplication this wrong entry could also block the real series, because it already occupied the same position slot.Compare with
<in the approximate scan and in the MESSI L2 leaf scan, so a distance that is not strictly better than the current k-th is never inserted.