Random forest on working on STM32N6570-DK

[chanchal94]

I am working on STM32N6570-DK. If I want to deploy random forest model on this MCU, what is the procedure. As most of the time I was using STM32 Cube AI studio to generate IDE project and work on it . But , AI studio doesn’t support tree based models. So, what is the standard procedure for deploying random forest or tree based models on this MCU for real time detection ? 
 

Kindly guide

Thanks & Regards

Chanchal 

[Julian E.]

Hi ​@chanchal94,

 

Please look at the layer support here:

Home - ST Edge AI Developer Cloud

 

My understanding is that some sklearn layers or algorithms are supported via ONNX export.

You will most likely find them in the Non NPU specific layers, meaning that you will not be able to use the --st-neural-art option.

 

Random forest on N6 seems a bit overkill, did you consider using NanoEdge AI Studio?

This AutoML tool will look for ML algorithms based on your data (RandomForest, SVM, MLP and more) and also preprocessing and give you a ready to deploy C library.

This is very easy to deploy and run on all Cortex M microcontrollers.

 

have a good day,

Julian