Convert model with input shape NCHW or NHWC to HWCN for inference

[Besma_Guesmi]

Hi

I have a model in PyTorch and TensorFlow frameworks. My expectation is to convert the model to accept an input shape [HWCN]. Is it possible to convert it using a model optimizer or any other method?

Thanks

[IntelSupport]

Hi Besma_Guesmi,

Thanks for reaching out.

It is possible to change the layout by including additional input pre-processing into the converted model using the --layout parameter. You can refer to When to specify Layout and How to Change Layout of a Model Inputs and Outputs documentation.

However, the HWCN layout value is not in the list of the Layouts that the OpenVINO inference engine support.

Regards,

Aznie

[IntelSupport]

Hi Besma_Guesmi,

This thread will no longer be monitored since we have provided information. If you need any additional information from Intel, please submit a new question.

Regards,

Aznie