TinyML Cookbook - Second Edition: Combine machine learning with microcontrollers to solve real-world problems

Over 70 recipes to help you develop smart applications on Arduino Nano 33 BLE Sense, Raspberry Pi Pico, and SparkFun RedBoard Artemis Nano using the power of machine learning

Purchase of the print or Kindle book includes a free eBook in PDF format.

Key Features:

• Train, optimize, and deploy ML models using TensorFlow Lite and Edge Impulse
• Get to grips with embedded platforms like Arm Mbed OS and Zephyr OS and peripherals like GPIO and I2C
• Explore cutting-edge technologies, such as on-device training for updating models without data leaving the device

Book Description:

Discover the incredible world of tiny Machine Learning (tinyML) and create smart projects using real-world data sensors with Arduino Nano 33 BLE Sense, Raspberry Pi Pico, and SparkFun RedBoard Artemis Nano.

You'll learn the unique constraints of on-device ML and how to work with embedded platforms like Arm Mbed OS. TinyML Cookbook, Second Edition, will show you how to implement end-to-end smart applications in different scenarios using the three "V" sensors (Voice, Vision, and Vibration). You'll train custom models from weather prediction to real-time speech recognition using TensorFlow Lite and Edge Impulse. Expert tips will help you squeeze ML models into tight memory budgets and accelerate performance using CMSIS-DSP. Finally, you'll learn advanced techniques like on-device learning, deploying scikit-learn models, and power optimization.

This edition includes new recipes featuring an LSTM neural network to recognize music genres and the Faster-Objects-More-Objects (FOMO) algorithm for detecting objects in a scene. These will help you stay up to date with the latest developments in the tinyML community.

Finally, take your tinyML solutions to the next level with microTVM, microNPU, and on-device learning. This book will give you the knowledge to make the most of your microcontroller and create unique projects with tinyML!

What You Will Learn:

• Understand the microcontroller programming fundamentals
• Work with real-world sensors, such as the microphone, camera, and accelerometer
• Run on-device ML with TensorFlow Lite for Microcontrollers
• Implement an app that responds to human voice with Edge Impulse
• Leverage transfer learning with FOMO and Keras
• Squeeze ML models into tight memory with quantization and other optimization methods
• Create gesture-recognition and music genre classifier apps with the Raspberry Pi Pico
• Design a CIFAR-10 model for memory-constrained microcontrollers

Who this book is for:

This book is ideal for machine learning engineers or data scientists looking to build embedded/edge ML applications and IoT developers who want to add machine learning capabilities to their devices. If you're an engineer, student, or hobbyist interested in exploring tinyML, then this book is your perfect companion.

Basic familiarity with C/C++ and Python programming is a prerequisite; however, no prior knowledge of microcontrollers is necessary to get started with this book.

超過70個食譜，幫助您使用機器學習的力量，在Arduino Nano 33 BLE Sense、Raspberry Pi Pico和SparkFun RedBoard Artemis Nano上開發智能應用程式。

購買印刷版或Kindle書籍，將包含一本免費的PDF電子書。

主要特點：
- 使用TensorFlow Lite和Edge Impulse訓練、優化和部署機器學習模型
- 熟悉像Arm Mbed OS和Zephyr OS這樣的嵌入式平台，以及像GPIO和I2C這樣的外設
- 探索尖端技術，例如在設備上進行訓練，以便在不將數據離開設備的情況下更新模型

書籍描述：
探索微小機器學習（tinyML）的令人難以置信的世界，並使用Arduino Nano 33 BLE Sense、Raspberry Pi Pico和SparkFun RedBoard Artemis Nano上的真實世界數據傳感器創建智能項目。

您將學習設備上機器學習的獨特限制，以及如何使用Arm Mbed OS等嵌入式平台。《TinyML Cookbook，第二版》將向您展示如何在不同情境中使用三個“V”傳感器（聲音、視覺和振動）實施端到端的智能應用程式。您將使用TensorFlow Lite和Edge Impulse從天氣預測到實時語音識別訓練自定義模型。專家提示將幫助您將機器學習模型壓縮到有限的記憶體預算中，並使用CMSIS-DSP加速性能。最後，您將學習設備上學習、部署scikit-learn模型和功耗優化等高級技術。

本版新增了使用LSTM神經網絡識別音樂類型和使用Faster-Objects-More-Objects（FOMO）算法檢測場景中物體的食譜。這些將幫助您跟上微小機器學習社區的最新發展。

最後，通過microTVM、microNPU和設備上學習將您的微小機器學習解決方案提升到更高水平。本書將為您提供充分利用微控制器並創建獨特項目的知識！

學到什麼：
- 理解微控制器編程基礎知識
- 使用麥克風、攝像頭和加速度計等真實世界傳感器
- 使用TensorFlow Lite for Microcontrollers在設備上運行機器學習
- 使用Edge Impulse實現對人聲的應答應用程式
- 利用FOMO和Keras進行轉移學習
- 使用量化和其他優化方法將機器學習模型壓縮到有限的記憶體中
- 使用Raspberry Pi Pico創建手勢識別和音樂類型分類器應用程式
- 為記憶體受限的微控制器設計CIFAR-10模型

本書適合機器學習工程師或數據科學家，他們希望構建嵌入式/邊緣機器學習應用程式，以及希望將機器學習功能添加到其設備的物聯網開發人員。如果您是工程師、學生或愛好者，對探索微小機器學習感興趣，那麼本書是您的完美伴侶。

基本熟悉C/C++和Python編程是先決條件；然而，不需要事先了解微控制器即可開始閱讀本書。