Integrate Automated QDQ placement tool - part 2.3

[willg-nv]

What does this PR do?

This PR implement RegionInspect tool. This tool could be used to visualize the regions parititioned by RegionSearch classes. This tool could be used to analyze if the partitioned regions match the fusion patterns.

Overview: ?

Usage

# Add a code snippet demonstrating how to use this

Testing

Before your PR is "Ready for review"

• Make sure you read and follow Contributor guidelines and your commits are signed.
• Is this change backward compatible?: Yes
• Did you write any new necessary tests?: Yes
• Did you add or update any necessary documentation?: No, document update is in Part 4.
• Did you update Changelog?: No, CHANGELOG will be updated when all changes are ready.

Additional Information

Summary by CodeRabbit

New Features

• Added a region inspection tool for ONNX models. Analyzes model structure and generates detailed reports including region statistics, hierarchical relationships, node coverage metrics, and size distribution analysis. Available through a command-line interface with configurable parameters.

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[coderabbitai]

Important

Review skipped

Auto incremental reviews are disabled on this repository.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

• 🔍 Trigger a full review

📝 Walkthrough

Walkthrough

Introduces a new module that provides region search inspection capabilities for ONNX models. The tool loads models, converts them to GraphSurgeon graphs, executes a two-phase region search, collects and filters regions, generates statistics, and exposes both programmatic and CLI interfaces.

Changes

Cohort / File(s)

Summary

Region Inspection Tool modelopt/onnx/quantization/autotune/region_inspect.py

New module (203 lines) implementing ONNX model region search inspection. Provides inspect_region_search() function for programmatic use and main() CLI entry point. Executes two-phase region search (partitioning and refinement), logs hierarchical relationships and per-region metrics, generates summary statistics including region type counts and coverage percentages, and produces size distribution histograms for LEAF regions.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Title check	⚠️ Warning	The title references an automated QDQ placement tool integration but the PR primarily introduces RegionInspect, a region search inspection visualization tool.	Revise the title to accurately reflect that this PR adds RegionInspect tool for visualizing and analyzing regions from RegionSearch, such as: 'Add RegionInspect tool for region search visualization and analysis'.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 2

🤖 Fix all issues with AI agents

In `@modelopt/onnx/quantization/autotune/region_inspect.py`:
- Around line 189-196: The log message is misleading because
inspect_region_search returns all regions but the logger.info says "top-level
regions"; update the logger.info call that logs len(regions) (the variable
assigned from inspect_region_search) to accurately describe the content (e.g.,
"regions discovered" or "all regions discovered") so the message reflects that
inspect_region_search returns all_regions rather than only top-level regions.
- Around line 47-57: Update the docstring in region_inspect.py so the API name
matches the implementation: replace references to search() with search_regions()
(and any nearby mention of "Two-Phase Search: Runs automatically via
CombinedRegionSearch.search()" to "CombinedRegionSearch.search_regions()") and
scan the surrounding text for any other mismatched mentions of search() to
ensure all documentation refers to the actual method
CombinedRegionSearch.search_regions().

[gcunhase]

@willg-nv can we add a unittest for this file? Thanks.

[willg-nv]

@willg-nv can we add a unittest for this file? Thanks.

Yes, please check the latest commit.

[gcunhase]

LGTM, deferring final approval to @ajrasane, thanks.