The example for quantization doesn't work

[mRaskovic]

Describe the bug The example from the readme file for model quantization doesn't work. Location: https://github.com/microsoft/onnxruntime/tree/master/onnxruntime/python/tools/quantization/E2E_example_model/image_classification/cpu

It crashes for the command: python run.py --input_model mobilenetv2-7.onnx --output_model mobilenetv2-7.quant.onnx --calibrate_dataset ./test_images/

The error I'm getting:

But it works properly for Resnet: python run.py --input_model .\resnet50-v1-13.onnx --output_model \resnet50-v1-13.quant.onnx --calibrate_dataset ./test_images/

Urgency Not urgent

System information

• OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Windows 10
• ONNX Runtime installed from (source or binary): pip install
• ONNX Runtime version: 1.8.1
• Python version: 3.7.0

To Reproduce Clone the repository and run the command from the readme file for the specified script.

Expected behavior It should perform the network quantization and compare inference times for FP32 and the quantized versions. The same way it works for resnet.

[yufenglee]

This bug was fixed in master #8788. Please try our nightly-build or older version 1.7.

[stale[bot]]

This issue has been automatically marked as stale due to inactivity and will be closed in 7 days if no further activity occurs. If further support is needed, please provide an update and/or more details.