NVIDIA / NVIDIA/raft

[FEA] lap_kernels.cu numeric_limits/infinity

Open
#96 4 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

inactive-30d inactive-90d
Dominant language
Cuda
Stars
1k
Forks
251
Avg merge
1d 8h
Merged PRs (30d)
31

Description

Once cuda 10.2 is the minimum supported version of cuda, we could modify lap_kernels.cu to use cuda::std::numeric_limits<weight_t>::max() instead of passing infinity into the kernels.

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

Start in lap_kernels.cu and inspect where infinity is passed into the kernels. Confirm the CUDA 10.2 minimum-version assumption and run the relevant kernel tests or existing project test suite; done means those calls use cuda::std::numeric_limits<weight_t>::max() without regressions.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
backend
Issue type
Feature
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Clearly specified
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.