[OMNIML-2244] Add E2E example for mixed precision quantization and ONNX export

[ajrasane]

What does this PR do?

Type of change:
New Feature

Overview:

• Enable ONNX export for auto quantized models
• Update documentation and changelog

Usage

python torch_quant_to_onnx.py --quantize_mode=auto \
	--onnx_save_path=./vit_base_patch16_224.nvfp4_fp8.onnx \
	--calibration_data_size 64 \
	--auto_quantization_formats NVFP4_AWQ_LITE_CFG FP8_DEFAULT_CFG \
	--batch_size 128

Testing

python evaluate.py --onnx_path=vit_base_patch16_224.nvfp4_fp8.onnx \
	--model_name=vit_base_patch16_224 \
	--results_path=./results.txt \
	--batch_size 128

Accuracy results

The top1 accuracy of the model is 84.15%
The top5 accuracy of the model is 97.396%

Reference accuracy for fp16

The top1 accuracy of the model is 85.102%
The top5 accuracy of the model is 97.526%

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?: No
• Did you add or update any necessary documentation?: Yes
• Did you update Changelog?: Yes

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 74.46%. Comparing base (53a2dde) to head (6c9f7b1).
⚠️ Report is 5 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #656      +/-   ##
==========================================
- Coverage   74.50%   74.46%   -0.05%     
==========================================
  Files         183      183              
  Lines       18400    18415      +15     
==========================================
+ Hits        13709    13712       +3     
- Misses       4691     4703      +12     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[gcunhase]

We should add perf and accuracy numbers for the baseline and quantized models in the README file as well.

[vishalpandya1990]

A basic query: Is onnx_ptq the right place of "PyTorch PTQ => ONNX export" examples? I was under the impression that onnx_ptq exemplifies PTQ techniques for input ONNX models.

[gcunhase]

A basic query: Is onnx_ptq the right place of "PyTorch PTQ => ONNX export" examples? I was under the impression that onnx_ptq exemplifies PTQ techniques for input ONNX models.

Right now, onnx_ptq includes PyTorch PTQ => ONNX examples as well, but I agree that it might be better to separate those 2 workflows. WDYT @ajrasane?

[kevalmorabia97]

You'll need to change test from onnx_ptq to torch_onnx as well in https://github.com/NVIDIA/Model-Optimizer/blob/main/.github/workflows/example_tests.yml#L119-L147 since we dont have any other onnx_ptq test remaining after this change

[gcunhase]

NVFP4 Quantized is NVFP4 + FP16-AutoCast? Same question for Auto Quantized.

[ajrasane]

Yes

[gcunhase]

Thanks for dividing the examples into 2 folders.

I posted a couple more comments and we'd still need to add the perf numbers to show the accuracy-runtime trade-offs.

Approving for now.