Model validation in MCU runtime on STM32N6

[brianon]

Hi,

I am currently working with the nucleo-n657x0-q and trying to validate a tflite model on it. While it works in the neural-ART runtime I cant seem to get a validation program working for the MCU runtime. For the neural-ART runtime I followed the NPU_Validation guide from the 2.1 ST Edge projects package, this worked without problem.

For the MCU runtime, my first approach was to follow the guide mentioned above but omit the --st-neural-art flag for the ST cli generate command, when running the n6_loader.py script this results in a "Compilation failed". Attached are the logs produced with the error.

My second approach was to use the cubeMX to generate a validation program. The main issue here is that each time that I enable validation in the X-cube-ai package the software freezes 99% of the time. I have been able to work around this issue by first creating a template program in MX and then changing to validation and regenerating the code. What I am stuck on now is that I do not know how to proceed with the program. Using validation on host in the .ioc only results in the following errors:

E200(ValidationError): TARGET: Unable to bind the ST.AI runtime with "network" c-model: []  E801(HwIOError): could not open port 'COM3': PermissionError(13, 'Access is denied.', None, 5)

Why am I getting these errors and is there a different approach that I am missing? Attached is also the MX generated project.

Thanks in advance,

Brian.

[Julian E.]

Hello @brianon,

Thank you for the information, I will try to replicate.

Have a good day,

Julian

[Julian E.]

Hello @brianon,

So, I learned that it is indeed not currently supported.

It will be added in the next update, ST Edge AI Core 2.2. 

I don't have the exact date, but it should come out in around 2 weeks.

Have a good day,

Julian

[brianon]

Hi @Julian E.,

That is unfortunate. For clarities sake, does this extend to all x-cube-ai functionality on the MCU runtime for the stm32n6 board or is it only validation that is not yet supported?

[Julian E.]

Hello @brianon,

To my knowledge, it is only the validation that is not yet supported.

Have a good day,

Julian