Adapt "How to build an AI app from scratch on STM32N6570-DK" to NUCLEO-N657X0-Q

[sop]

I am trying to make a minimal working example of a neural network running on the NUCLEO-N657X0-Q. To achieve this I followed this guide for the STM32N6570-DK on my own NUCLEO-N657X0-Q board. (https://community.st.com/t5/stm32-mcus/how-to-build-an-ai-application-from-scratch-on-the-stm32n6570-dk/ta-p/825591#!/825591)

However, CubeMX did not initialize the required code lines noted in the guide under "Important note 2".

Guide's screenshot:

My screenshot:

I did small address adaptions for the NUCLEO-N657X0-Q

• LEDs: PG0, PG8
• EXT_SMPS_MODE: PB12

I will attach the project below. Did I make a mistake or is CubeMX not able to configure the NUCLEO-N657X0-Q in the same way it supports the STM32N6570-DK yet? I also tried adding the lines from the screenshot of the guide myself, but it didn’t change the results and I assume there is more missing than is visible in the guide's screenshot.  I also assume the missing RIF and RIMC code lines are the main reason why the application didn’t work.

I still wrote all necessary files to flash, but the result is that I don’t get any messages from the serial port. Only the FSBL LED turns on and off again. The LED toggled in the Appli remains turned off.

I appreciate any help on this issue.

Is there another minimal AI example out for the Nucleo board, not the Discovery Kit?

Used SW:

CubeIDE 1.19.0

CubeMX 6.15.0

CubeProgrammer 2.20.0

X-Cube-AI 10.2.0

STM32Cube MCU package N6: 1.2.0

Used HW:

NUCLEO-N657X0-Q

OTP124=0x00008000

 

 

[Julian E.]

Hello @sop,

 

I am doing the same thing to write an article on the Nucleo.

I am struggling with strange behaviors that I think come from CubeMX.

 

As you pointed out, sometimes it does not initialize the code for the RIF correctly, or the RAM.

 

I managed to create a project that works on the Nucleo, but I want to test it further before giving out the steps as I struggles to succeed 100% of the time. (I may do mistakes sometimes, but I feel like I have more issues when I use the standalone CubeMX than the one inside CubeIDE)

 

If I succeed replicating it, I should write the article quickly, as it is almost identical to the DK one.

 

Have a good day,
Julian

[sop]

For completeness, the new guide for the Nucleo-N6 can be found here:
https://community.st.com/t5/stm32-mcus/how-to-build-an-ai-application-from-scratch-on-the-nucleo-n657x0/ta-p/828502#!/828502