Part Number: MSPM0G5187
Hi,
Let me confirm about below.
I'm trying to perform following tutorial.
https://dev.ti.com/tirex/explore/node?node=A__AKCnvqDed-Plz2JO5Umb3Q__MSPM0-SDK__a3PaaoK__LATEST
Then, when I perform "torch.onnx.export(modelInt8, inData, "trained_model.onnx", input_names=["input"])" of "3. Train Your Own Model (Simple Native Pytorch Quantization)" in above URL, I got following error.
* torch.onnx.errors.UnsupportedOperatorError: Exporting the operator 'quantized::batch_norm2d' to ONNX opset version 17 is not supported.
According to following page, opset version of onnx==1.18.0 should be "23", however error show "17".
This point is also strange. Did you experienced same error as shown above ?
onnx/docs/Versioning.md at main · onnx/onnx
Best Regards,
