May I ask, the model uses the torch.mul() operator, and when the model is saved in ONNX format and uploaded to ST Edge AI, it fails to parse. Ho

[mƎALLEm]

Hello @shua and welcome to the ST community,

Please review How to write your question to maximize your chances to find a solution

Especially this point:

I invite you to edit your post accordingly.

Thank you for your cooperation.

[Julian E.]

Hi @shua,

It most likely depends on your model.

For non-accelerated hardware, you can find the list of supported layers here (for onnx), and in your case mul:

https://stedgeai-dc.st.com/assets/embedded-docs/supported_ops_onnx.html#mul

If you are using a NPU (n6 for example): 

https://stedgeai-dc.st.com/assets/embedded-docs/stneuralart_operator_support.html#title

 Make sure to respect the conditions listed

Have a good day,

Julian