pytorch / pytorch/executorch

Add a xnnpack partitioner flag to dump XNNGraph AOT

Open
#7,591 3 comments 1 reaction 1 assignee View on GitHub

@digantdesai is already working on this.

Since Jan 10, 2025.

actionable module: xnnpack triaged
Dominant language
Python
Stars
5k
Forks
1.2k
Avg merge
2d 10h
Merged PRs (30d)
581

Description

🚀 The feature, motivation and pitch

This is essential for any deeper debugging beyond the surface level. This is something we have been asked multiple times for anyone doing serious model development work.

Alternatives

None.

Additional context

When serializing XNNGraph in the partitioner, we already have this information as json. This optional flag will dump it in a file, for further debugging.

RFC (Optional)

None

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.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.