How to integrate Cube AI Studio generated Neural-ART code into an existing project?

[emmanuel_]

 

Dear all,

For the STM32N6570-DK kit, is there a simple tutorial that shows the workflow for integrating a model generated by STM32Cube AI Studio into a project?

A simple example would be: if an image is detected, we could turn on an LED on the kit.

A guide would be very welcome. Many thanks

Cheers

Emmanuel

[Julian E.]

Hi @emmanuel_ ,

 

In STM32cube AI Studio, you can generate 2 things:

• Only the C code of the model
• A template project 

The template project on seems to be what you request.

In the current version it doesn't do anything else than running an inference.

 

In the next version (end of april) we will also print the inference time via serial.

 

We are working on a tutorial to explain how to recreate manually the template project from a cubeMX blank project, generating the C files of the model, copying them in the project and recreating the simple application.

Do you think this is helpful? please let us know what you think, so that we can adapt our tutorial.

 

In the meantime, if you generate the project, I think that what you are looking for is in AI\App\x-cube-ai.c

The main.c call the function in the other file in the while(1) loop, nothing more.

 

Have a good day,

Julian

[ayaaaa]

Hello , can you please include in the tutorial the correct debugging method for the generated project ? that would be helpful 

and also when can we expect this tutorial ?

[emmanuel_]

Hi @Julian E. ,

Thank you very much for your response.

This tutorial would be very helpful, especially if it clearly shows the steps to follow in order to successfully complete the project. 

The simplest tutorials are sometimes the most effective !

 

Cheers

 

Emmanuel