Any body played w/ any of this?
On a RPi or STM32 Nucleo brd?
https://www.dlology.com/blog/how-to-run-deep-learning-model-on-microcontroller-with-cmsis-nn/
@denzuko … is that you? 
Actually, @thespacemaker and @mrcity and @mrhavens and @dougemes might be interested
I’m not certain but @otto1, @larrydag and @semaphore1999
I was doing a quick look through the Events Cal and found Mark’s class:
It’s fairly easy with a Raspberry Pi but then again anything running python jupyter and linux would do.
One just boots up then opens a terminal then runs this to install:
sudo apt-get update && sudo apt-get install -y python3-matplotlib python3-scipy
sudo pip3 install --upgrade pip
sudo pip3 install jupyter karas
Next add your .ipynb files to a project directory, change into that directory and start jupyter
To start python jupiter one would type
jupyter-notebook
From there one would have a browser window pop up that should have all your jupyter notebooks loaded.
The main part to take away here is one is just running python modules (scipy, matplotlib, karas, numpy) and any C code to execute the ML models.
1 Like
Catching up on techie newsletters, etc.
Not on ARM.
2 Likes
That USB Accelerator card looks interested. Kind of reminds me of the movidius sticks we already have in VCC’s committee area.