llvm / llvm/torch-mlir

[RFC] Working on aten.norm.ScalarOpt_dim

Open
#1,002 7 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
1.9k
Forks
736
Avg merge
5d 22h
Merged PRs (30d)
15

Description

Just trying to confirm that lib/Dialect/Torch/Transforms/DecomposeComplexOps.cpp is the right place for the decomposition of this op. Would like a few starting hints at how to do a decomposition since this is my first one, and what exactly am I decomposing this op into? Here is my PR: https://github.com/llvm/torch-mlir/pull/949/files. And the logic of the op: https://www.google.com/search?q=p+norm&source=lnms&tbm=isch&sa=X&ved=2ahUKEwjjyNS3wdj4AhV6HzQIHQg_B4YQ_AUoAXoECAIQAw&biw=1707&bih=928&dpr=1.5#imgrc=03A1IiA6Bl7BNM. I'm going to need addition for the sum (AddFOp), exponents (PowFOp), and absolute value (AbsOp).

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 with lib/Dialect/Torch/Transforms/DecomposeComplexOps.cpp and review the proposed changes in PR #949. Confirm the intended decomposition of aten.norm.ScalarOpt_dim and how AddFOp, PowFOp, and AbsOp fit into it; the work is done when the decomposition approach is agreed and implemented there.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp, pytorch
Domain
compilers
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.