kokkos / kokkos/kokkos.github.io

Direct Measurement Numerical Accuracy

Open
#94 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

KUG2023 LightningTalk User
Dominant language
HTML
Stars
3
Forks
27
Avg merge
2d 25m
Merged PRs (30d)
11

Description

Shaman uses Error Free Transforms to explicitly evaluate order 1 rounding error of IEEE floating point. It works as a library and it just needs the user to replace their scalar types by Shaman's. We will soon adapt it to integrate well with Kokkos and be able to assess numerical accuracy of GPU kernels.
The long term goal is to ease the move to cheaper and faster datatypes introduced for AI computations.

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

The issue names Shaman, Error Free Transforms, IEEE floating point, Kokkos, and GPU kernels, but no repository file, test, or entry point. Start by clarifying the intended website or integration change and its acceptance criteria; completion cannot be determined from the current description.

Written by the indexing model from the issue text.

Assessment

Domain
performance
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.