NVIDIA / NVIDIA/cuvs

MG KMeans initialization emits unnecessary warnings

Open
#2,319 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

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

Description

The multi-GPU KMeans initialization and more specifically the reclustering step of the KMeans|| initialization method emits unnecessary/confusing warnings when used in MNMG mode (RAFT communicator).

This is caused by the fact that the reclustering step of the KMeans|| initialization method calls the public MG KMeans API interface that handles SG/MG dispatch.

The warning :

Multi-GPU handle detected on single-GPU kmeans::fit() entry; falling back to single-GPU. Use cuvs::cluster::kmeans::mg::fit(...) for multi-GPU.

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 in cpp/src/cluster/detail/kmeans_mg_distributed_init.cuh at the reclustering call around line 444, where KMeans|| initialization uses the public MG KMeans API interface. Trace the SG/MG dispatch in that path and verify MNMG initialization no longer emits the single-GPU fallback warning while retaining the correct multi-GPU behavior.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp, machine-learning
Domain
distributed-systems, machine-learning
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
56/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.