[hipDNN] Test cpu graph executor for batch norm inference variance

[bghimireamd]

Motivation

• Added missing tests for BatchnormFwdInferenceWithVariancePlan.

• Fixed incomplete validation in BatchnormInferenceNodeVarianceExt which was missing the check for the epsilon tensor, ensuring standard validation catches missing inputs.

Test Plan

Verify CPU Graph Executor tests (Execution, Plan Construction, Applicability)
./bin/hipdnn_test_sdk_tests --gtest_filter="*BatchnormFwd*VariancePlan*"

Verify Frontend Validation tests (Missing Epsilon check)
./bin/hipdnn_frontend_tests --gtest_filter="*BatchnormInferenceNodeVarianceExt*"

Test Result

Submission Checklist

• Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

[Copilot]

Pull request overview

This PR enhances test coverage and validation for the batch normalization inference with variance feature. It adds comprehensive CPU graph executor tests and fixes a validation gap in the frontend node.

Changes:

• Added three new tests (ExecutePlan, PlanConstruction, IsApplicable) for BatchnormFwdInferenceWithVariancePlan in the test SDK
• Fixed missing epsilon tensor validation in BatchnormInferenceNodeVarianceExt::pre_validate_node
• Extended test utilities (BatchnormGraphUtils and BatchnormTensorBundles) to support variance-based batch normalization tests

Reviewed changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
TestBatchnormFwdInferenceWithVariancePlan.cpp	Adds comprehensive tests for plan execution, construction, and applicability checks
BatchnormTensorBundles.hpp	Introduces BatchnormFwdWithVarianceTensorBundle struct for test tensor management
BatchnormGraphUtils.hpp	Adds buildBatchnormFwdInferenceWithVarianceGraph helper function for test graph creation
BatchnormFwdInferenceWithVariancePlan.hpp	Adds epsilon tensor existence validation to plan builder
TestBatchnormInferenceNodeVarianceExt.cpp	Adds test case for missing epsilon tensor validation
BatchnormInferenceNodeVarianceExt.hpp	Adds epsilon tensor validation check to prevent incomplete node configuration

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 6 out of 6 changed files in this pull request and generated 3 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[BrianHarrisonAMD]

Looks good assuming tests pass.

I think it would be worth double checking the value init, and epsilon pass-by-value checks in frontend, but I don't think it's blocking.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 90.00000% with 2 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...rontend/node/BatchnormInferenceNodeVarianceExt.hpp	80.00%	0 Missing and 1 partial ⚠️
...executor/BatchnormFwdInferenceWithVariancePlan.hpp	90.91%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##           develop    #3959       +/-   ##
============================================
+ Coverage    49.42%   81.13%   +31.71%     
============================================
  Files          135      156       +21     
  Lines        33698    15865    -17833     
  Branches      4436     1993     -2443     
============================================
- Hits         16654    12871     -3783     
+ Misses       15865     2216    -13649     
+ Partials      1179      778      -401     

Flag	Coverage Δ
hipDNN	81.13% <90.00%> (?)
hipFFT	?
rocFFT	?

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
...dnn/frontend/include/hipdnn_frontend/Utilities.hpp	92.57% <100.00%> (ø)
...rontend/node/BatchnormInferenceNodeVarianceExt.hpp	87.95% <80.00%> (ø)
...executor/BatchnormFwdInferenceWithVariancePlan.hpp	71.26% <90.91%> (ø)

... and 288 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[adickin-amd]

Issuing override merge. Lots of slowdowns on windows gfx1151. This is being worked on and this PR doesnt need to be adding to the queue times.