onnx / onnx/models

Add self-hosted machine with VNNI support in GitHub Action CI

Open
#574 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

CI enhancement
Dominant language
Jupyter Notebook
Stars
9.8k
Forks
1.6k
PR merge metrics
No merged PRs in 30d

Description

Feature Request

What is the problem that this feature solves?

The test data for quantized models will be generated from machines with VNNI support. General machines without VNNI support cannot repro the correct result and currently CIs do not provide VNNI support. See issue: https://github.com/onnx/models/issues/568. Currently these data cannot be verified by existing CIs. Thus, as a workaround, the reviewer needs to run ORT inference in a local VNNI machine.

Describe the feature

Add self-hosted machine with VNNI support in GitHub Action CI and so the CI can help to verify those data generated from machines with VNNI support.

Relevant Model Zoo feature areas

Which area in the Model Zoo infrastructure does this impact?
Feature Area (e.g. CI, file organization, model storage, other): CI

Notes

Any additional information

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 by reviewing the repository's GitHub Action CI configuration and the linked onnx/models issue 568 to understand how quantized-model data is currently verified. Determine how a self-hosted VNNI machine would be registered and targeted, and consider the work done when CI can verify data generated on VNNI-capable hardware.

Written by the indexing model from the issue text.

Assessment

Tech stack
github-actions
Domain
ci-cd, infrastructure
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
32/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.