NVIDIA / NVIDIA/cuvs

[FEA] Use "batched all-neighbors" for out of core CAGRA build on binary quantized vectors

Open
#1,200 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

feature request
Dominant language
Cuda
Stars
854
Forks
236
Avg merge
3d 3h
Merged PRs (30d)
62

Description

Build a CAGRA index with the BitwiseHamming metric by building the graph out of core. Use batched nn descent for this.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by locating the CAGRA out-of-core build and the batched nn descent implementation. Verify how the BitwiseHamming metric is handled, then determine the build path needed to construct the graph from binary quantized vectors. Done means a CAGRA index builds out of core with BitwiseHamming using batched nn descent.

Written by the indexing model from the issue text.

Assessment

Domain
performance, search
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.