cannot build application after generating C code model with St edge AI core CLI

[s319189]

I'm following this guide to deploy an AI model on a NUCLEO-H7S3L8, because appearently it is not possible to build and edge AI application through X-CUBE-AI on STM32CubeMX (because it does not generate network.h etc. but only app_x-cube-ai.c and app_x-cube-ai.h).

https://stedgeai-dc.st.com/assets/embedded-docs/stm32_how_to_run_a_model_locally.html

however, I'm having trouble on the build process wehre it is written 

"Build the test application"

where it is needed to link different libraries, because there is no indication on what to do precisely and where to insert the commands.

Could anybody provide a clearer and more precise guide to implement this step?

Thanks for the reply

[Julian E.]

Hi @s319189,

 

I would suggest to use the new tool STM32CubeAI studio and the ST Edge AI Core 3.0.

https://community.st.com/t5/developer-news/introducing-stm32cubeai-studio/bc-p/880860#M368

 

Have a good day,

Julian

 

[s319189]

I've tried to use it, but when I try to modify my project through CubeMX and generate new code I cannot view/modify .h files such as main.h inside STM32CubeIDE because the project generated by Cube AI studio  is not under the root 

[hamitiya]

Hello,

Generating under root for STM32CubeIDE project can cause conflicts if you have other projects in your subfolder.

I answered you in a previous ticket about the fact that you cannot see header files by default in your file tree, that does not mean that you can't edit it.

If you want to edit it, you can go to your main.c, unfold file, and double-click on your header file.

 

 

Best regards,

Yanis