Honey, I shrunk the TinyML
You might remember the movie from the 80’s when Wayne Szalinski shrunk his family. Back then, machine learning was also little. But I don’t want to talk about the past, I want to talk about the future, where machine learning is getting tiny. I will show how to train a model and convert it to use TinyML on a microcontroller. Which possibilities this could open up now and in the future. What are the possibilities to shrink a model with quantization and pruning, but also what are the limitations at the moment. In this presentation I show step by step how to create a model, train it and convert it as TinyML. I also show how to integrate the model and run it as C++ code on a microcontroller.