machine learning with tensorflow lite on arduino and ultra-low-power microcontrollers Warden, Pete creator Author Situnayake, Daniel Joint author text cc Mumbai O'Relly Media Inc. 2021 2020 monographic eng

print

xvi, 484 pages : illustrations ; 24 cm Deep learning networks are becoming smaller, with models as small as 14 kilobytes. This practical book, TinyML, combines deep learning and embedded systems to create small, portable devices. It provides a step-by-step guide for developers to create TinyML projects, including speech recognition, camera detection, and gesture response. The book also covers learning ML basics, using TensorFlow Lite for microcontrollers, and optimizing latency and energy usage. Introduction -- Getting started -- Getting up to speed on machine learning -- The "Hello world" of TinyML : building and training a model -- The "Hello world" of TinyML : building an application -- The "Hello world" of TinyML : deploying to microcontrollers -- Wake-word detection : building an application -- Wake-word detection : training a model -- Person detection : building an application -- Person detection : training a model -- Magic wand : building an application -- Magic wand : training a model -- TensorFlow lite for microcontrollers -- Designing your own TinyML applications -- Optimizing latency -- Optimizing energy usage -- Optimizing model and binary size -- Debugging -- Porting models from TensorFlow to TensorFlow Lite -- Privacy, security, and deployment -- Learning more. Pete Warden and Daniel Situnayake. Includes bibliographical references and index. Machine learning Signal processing Digital techniques Microcontrollers 006.31 WAR/T 9789352139606 (PBK) IN-BhIIT 210714 20260310154943.0 TB12444