microsoft / microsoft/winml-cli

P0-FEATURE-010: Graph Optimizer - More Advanced Optimizations

Open
#106 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

feature scale graph-optimizer P1 triaged
Dominant language
Python
Stars
40
Forks
11
Avg merge
1d 8h
Merged PRs (30d)
50

Description

Summary

Implement advanced graph optimizations including Cast movement and NonZero elimination.

Context

  • Priority: P0.17
  • Category: Feature Scale - Graph Optimizer

Scope

  • Move Cast around non-compute ops (reduce precision conversion overhead)
  • NonZero elimination (remove unnecessary NonZero ops)
  • Advanced fusion patterns
  • EP-friendly graph transformations

Acceptance Criteria

  • Cast movement optimization implemented
  • NonZero elimination working
  • Advanced fusion patterns applied
  • Validated on P0 models with no regressions

References

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 by reading plans/release/0315_release_plan/feature-scale.md to identify the graph optimizer entry points, supported patterns, and P0 validation models. Define and validate Cast movement, NonZero elimination, and advanced fusion patterns, then run the referenced P0 model checks and confirm there are no regressions.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.