Shawn Hymel

Glad you found it helpful! The .h5 model is a Keras model, so once you load it into your program, you can use model.predict() ([https://www.tensorflow.org/api_docs/python/tf/keras/Model#predict](https://www.tensorflow.org/api_docs/python/tf/keras/Model#predict)) to make predictions. From my...

I'll keep it in mind as a to-do, but I don't know if I'll have time for it right now. Please note that I'm working on a newer/better version of...

If you print `output_data`, what do you see? You should see a set (list) of probabilities for each of your classes.

I just checked, and the nucleo-l476-keyword-spotting project builds fine on STM32CubeIDE 1.6.0 for me. Which version of STM32CubeIDE are you using? Which project are you attempting to build? Did you...

I have not tried any of this with the SensorTile, so I don't know if it will work. Did you make any modifications to the code?

It looks like the build system can't find some of the libraries (such as the edge-impulse-sdk directory). If you're just copying and pasting code, that won't work. You need to...

There are a lot of changes you'll need to make, as none of my projects were created for the SensorTile, and the microphone on the SensorTile PDM (whereas I used...

I'm not sure if it will work--I have not tried to use a PDM microphone with this model nor have I tried running the model on a SensorTile. I recommend...

Hi @robertfchapman, Apologies for the delay. If I recall, the CM4 is considered a single footprint, which consists of the two connectors. The only way I know how to handle...

You will likely need to go through the process of setting up the board using CubeMX. The code that's generated from CubeMX only applies to one chip (or board).