NVIDIA / NVIDIA/cuvs

[FEA] Consolidate template parameters

Open
#370 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

We've had to add template parameters for both int64_t and int32 indices in the public api. We should consolidate these types to a single integral type where possible in the public api.

For functions like pairwise distances that benefit from working on int32 sized data, we should dispatch privately from int64_t to int32_t - and potentially tile the calculations for large inputs.

We should also examine consolidating the double and single precision float inputs : I believe that faiss only supports float32 data, and that might also be an option here

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

The issue names no files, tests, or entry points. Start by auditing the public API for int64_t/int32_t and float precision parameters, then trace the pairwise-distance paths. Done means the consolidation scope, private dispatch or tiling requirements, and supported public types are specified and validated.

Written by the indexing model from the issue text.

Assessment

Domain
backend-api-design, machine-learning, performance
Issue type
Refactor
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.