NVIDIA / NVIDIA/cuvs

[BUG] CAGRA extend operation does not support float16 data type

Open
#1,595 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

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

Description

Bug Description

CAGRA extend operation does not support float16 data type, while other CAGRA operations already support it.

Related Issues/PRs

  • #1419 - Initial request for float16 support in CAGRA
  • #1429 - Added float16 support for CAGRA build/search operations

Problem

PR #1429 added float16 support to CAGRA build and search operations, but the extend operation was not updated. This creates an inconsistency where users can build and search indices with float16, but cannot extend them.

Solution

Possible solution: https://github.com/TomaszGrabowskiPawlak/cuvs/pull/1/files

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 CAGRA’s extend implementation and comparing it with the float16 support added for build and search in PR #1429; the issue provides no file or test path. Done means extend accepts float16 data consistently with those operations, with coverage for the float16 extend case.

Written by the indexing model from the issue text.

Assessment

Domain
search
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.