pytorch / pytorch/executorch

Partition only user-annotated subgraphs

Open
#13,657 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

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

Description

🚀 The feature, motivation and pitch

For some advanced use cases, users may want to control which portions of the model are lowered to specific backends. This is possible by writing a custom partitioner (see https://docs.pytorch.org/executorch/main/compiler-delegate-and-partitioner.html#common-questions), but it would be nice to have a high-level API that works with to_edge_transform_and_lower to do this. The API design needs discussion, but it would be nice to not need to write a partitioner for this use case.

Alternatives

No response

Additional context

No response

RFC (Optional)

No response

cc @JacobSzwejbka @angelayi

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

Start with to_edge_transform_and_lower and the compiler delegate and partitioner documentation, especially the custom-partitioner common questions. Clarify the high-level API and how users mark subgraphs before implementing it; done means user-annotated subgraphs alone are lowered to the selected backend without requiring a custom partitioner.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
compilers, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.