Deployment of multiple Neural Networks on STM32n6

[ShiLiba]

Hello! Has anyone tried to deploy multiple neural networks on a stm32n6 at once? I'm developing an application which requires 5 NN. I created 5 .c libraries to deploy each NN on the NPU, I assume is because of the APIs that they overlap. So I'm curious if anyone has tried this and how it was solved.

My last  option is to create a bigger models that contains all 5 on onnx, then deploy it, idk if someone knows if this could work to. 

[Julian E. · Accepted solution]

Hi @ShiLiba,

I think that using a single model might be easier indeed.

To use multiple models you need to use the --name option when doing the generate, by default the model name is "network". If all your models are named "network" it will never compile because of multiple define references.

Then, you need to put all weights at different places in memory (by playing with the memorypool files and profiles) and to make sure that all 5 models fit in the code. Nothing fancy.

We don't have documentation about that specifically, but you should find some help about the general concept:

https://stedgeai-dc.st.com/assets/embedded-docs/index.html#neural-art-accelerator-target

Have a good day,

Julian