Skip to content

fix(index): compare Hamming IVF centroids during merge - #8132

Merged
Xuanwo merged 1 commit into
lance-format:mainfrom
jackye1995:codex/fix-hamming-centroid-merge
Aug 1, 2026
Merged

fix(index): compare Hamming IVF centroids during merge#8132
Xuanwo merged 1 commit into
lance-format:mainfrom
jackye1995:codex/fix-hamming-centroid-merge

Conversation

@jackye1995

Copy link
Copy Markdown
Contributor

Summary

  • compare UInt8 IVF centroid tensors exactly during distributed index merges
  • add regression coverage for matching and differing Hamming centroids

@github-actions github-actions Bot added A-index Vector index, linalg, tokenizer bug Something isn't working labels Aug 1, 2026

@lance-gatekeeper lance-gatekeeper Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Gate recommendation: approve. The base path rejects all multi-shard Hamming merges because the shared centroid comparator handles only floating-point tensors. Exact UInt8 equality at that shared invariant restores valid merges while continuing to reject incompatible shapes, types, and bytes, without changing floating-point tolerance behavior.

@codecov

codecov Bot commented Aug 1, 2026

Copy link
Copy Markdown

Codecov Report

❌ Patch coverage is 95.45455% with 1 line in your changes missing coverage. Please review.

Files with missing lines Patch % Lines
...lance-index/src/vector/distributed/index_merger.rs 95.45% 0 Missing and 1 partial ⚠️

📢 Thoughts on this report? Let us know!

@justinrmiller justinrmiller left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Might consider a generic fallback but lgtm

@Xuanwo
Xuanwo merged commit 6027511 into lance-format:main Aug 1, 2026
40 of 41 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

A-index Vector index, linalg, tokenizer bug Something isn't working

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants