TinyML: Machine Learning with TensorFlow Lite on Arduino and Ultra-Low-Power Microcontrollers

October 31, 2020
TinyML: Machine Learning with TensorFlow Lite on Arduino and Ultra-Low-Power Microcontrollers

Deep learning networks are getting smaller. Much smaller. TheGoogle Assistant team can detect words with a model just 14kilobytes in size—small enough to run on a microcontroller. Withthis practical book you'll enter the field of TinyML, where deeplearning and embedded systems combine to make astounding thingspossible with tiny devices.Pete Warden and Daniel Situnayake explain how you can trainmodels small enough to fit into any environment. Ideal for softwareand hardware developers who want to build embedded systems usingmachine learning, this guide walks you through creating a series ofTinyML projects, step-by-step. No machine learning ormicrocontroller experience is necessary.Build a speech recognizer, a camera that detects people, and amagic wand that responds to gesturesWork with Arduino and ultra-low-power microcontrollersLearn the essentials of ML and how to train your ownmodelsTrain models to understand audio, image, and accelerometerdataExplore TensorFlow Lite for Microcontrollers, Google's toolkitfor TinyMLDebug applications and provide safeguards for privacy andsecurityOptimize latency, energy usage, and model and binary size