ML on the Edge: Turning your RPi into an AI in under 45 minutes

This repository contains the materials and code for the tutorial given at SoFlo Dev Con 2022 by Misha Klopukh.

Description

In this workshop, we will make a Raspberry Pi-based image classifier from scratch. In just 45 minutes, we will:

• Generate a dataset from an image search;
• Fine-tune a neural network classifier in the cloud;
• Load our model onto a Raspberry Pi for inference;
• Create a real-time camera -> model -> display loop; and
• Demonstrate our new device in practice.

Materials

Everything in this tutorial is cheap and easy to get. To follow along, you'll need:

• A computer with access to the internet.
• A basic understanding of Python programming
• A google account (for Google Colaboratory)
• Optional: A Weights&Biases account (for run logging)
• A Raspberry Pi or equivalent (must be 64-bit)
• A camera module
• Some LED's

Instructions:

1. Run through the training notebook
   • You should run this notebook in the cloud with Google Colab
   • 
   • Make sure to select a GPU runtime
2. Follow these instructions to set up your Raspberry Pi
3. Make it your own!