NVIDIA / NVIDIA/raft

[ENH] mdspan-ify iterator-based API for `linalg::reduce_rows_by_keys`

Open
#925 2 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

0 - Backlog cpp improvement
Dominant language
Cuda
Stars
1k
Forks
251
Avg merge
1d 8h
Merged PRs (30d)
31

Description

Following #909, the API for linalg::reduce_rows_by_keys can take custom iterators, in order to save intermediate steps in ann_kmeans_balanced. But raw-pointer/iterator APIs are being deprecated in favor of mdspan.

We should provide appropriate helpers and types for iterator-based mdspan, and change this API accordingly. See discussions on the aforementioned PR.

Contributor guide

No contributing guide indexed for this repository

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

Read #909 and its discussions first, then trace linalg::reduce_rows_by_keys and its use in ann_kmeans_balanced. Identify the appropriate iterator-based mdspan helpers and types, and confirm done when the API no longer relies on the deprecated raw-pointer/iterator form while preserving that caller's needs.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
backend-api-design, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.