NVIDIA / NVIDIA/cuvs

[FEA] Support for half data types for balanced kmeans fit()

Open
#788 1 comment 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

kmeans::fit() only support int8_t(here) and float(here) data types right now.
Would be nice for it to support half data types (this will be able to reduce the memory usages of the batched knn build algorithm with ivfpq)

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 with the kmeans::fit() overloads in cpp/include/cuvs/cluster/kmeans.hpp, especially the linked int8_t and float definitions. Trace how the existing data types reach the balanced kmeans implementation and identify the validation coverage for fit(). Done means half data types are accepted by fit() and the relevant behavior is covered consistently.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
machine-learning, performance
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.