Book Image

TinyML Cookbook

By : Gian Marco Iodice
Book Image

TinyML Cookbook

By: Gian Marco Iodice

Overview of this book

This book explores TinyML, a fast-growing field at the unique intersection of machine learning and embedded systems to make AI ubiquitous with extremely low-powered devices such as microcontrollers. The TinyML Cookbook starts with a practical introduction to this multidisciplinary field to get you up to speed with some of the fundamentals for deploying intelligent applications on Arduino Nano 33 BLE Sense and Raspberry Pi Pico. As you progress, you’ll tackle various problems that you may encounter while prototyping microcontrollers, such as controlling the LED state with GPIO and a push-button, supplying power to microcontrollers with batteries, and more. Next, you’ll cover recipes relating to temperature, humidity, and the three “V” sensors (Voice, Vision, and Vibration) to gain the necessary skills to implement end-to-end smart applications in different scenarios. Later, you’ll learn best practices for building tiny models for memory-constrained microcontrollers. Finally, you’ll explore two of the most recent technologies, microTVM and microNPU that will help you step up your TinyML game. By the end of this book, you’ll be well-versed with best practices and machine learning frameworks to develop ML apps easily on microcontrollers and have a clear understanding of the key aspects to consider during the development phase.

Related Content you might be interested in

Current Title:

Small book image
TinyML Cookbook
Gian Marco Iodice
Apr, 2022 | 344 pages
Small book image
The Insider’s Guide to Arm Cortex-M Development
Pareena Verma | Jason Andrews | Zachary Lasiuk
Oct, 2022 | 276 pages
Small book image
Raspberry Pi Pico DIY Workshop
Srihari Yamanoor | Sai Yamanoor
May, 2022 | 376 pages
Small book image
DIY Microcontroller Projects for Hobbyists
Miguel Angel GarciaRuiz | Pedro Cesar Santana Mancilla
Jul, 2021 | 320 pages
Table of Contents (10 chapters)
Preface
Preface
Who this book is for
What this book covers
To get the most out of this book
Download the color images
Conventions used
Sections
Get in touch
Share Your Thoughts
1
Chapter 1: Getting Started with TinyML
Chapter 1: Getting Started with TinyML
Technical requirements
Introducing TinyML
Summary of DL
Learning the difference between power and energy
Programming microcontrollers
Presenting Arduino Nano 33 BLE Sense and Raspberry Pi Pico
Setting up Arduino Web Editor, TensorFlow, and Edge Impulse
Running a sketch on Arduino Nano and Raspberry Pi Pico
Join us on Discord!
Free Chapter
2
Chapter 2: Prototyping with Microcontrollers
Chapter 2: Prototyping with Microcontrollers
Technical requirements
Code debugging 101
Implementing an LED status indicator on the breadboard
Controlling an external LED with the GPIO
Turning an LED on and off with a push-button
Using interrupts to read the push-button state
Powering microcontrollers with batteries
3
Chapter 3: Building a Weather Station with TensorFlow Lite for Microcontrollers
Chapter 3: Building a Weather Station with TensorFlow Lite for Microcontrollers
Technical requirements
Importing weather data from WorldWeatherOnline
Preparing the dataset
Training the ML model with TF
Evaluating the model's effectiveness
Quantizing the model with the TFLite converter
Using the built-in temperature and humidity sensor on Arduino Nano
Using the DHT22 sensor with the Raspberry Pi Pico
Preparing the input features for the model inference
On-device inference with TFLu
4
Chapter 4: Voice Controlling LEDs with Edge Impulse
Chapter 4: Voice Controlling LEDs with Edge Impulse
Technical requirements
Acquiring audio data with a smartphone
Extracting MFCC features from audio samples
Designing and training a NN model
Tuning model performance with EON Tuner
Live classifications with a smartphone
Live classifications with the Arduino Nano
Continuous inferencing on the Arduino Nano
Building the circuit with the Raspberry Pi Pico to voice control LEDs
Audio sampling with ADC and timer interrupts on the Raspberry Pi Pico
5
Chapter 5: Indoor Scene Classification with TensorFlow Lite for Microcontrollers and the Arduino Nano
Chapter 5: Indoor Scene Classification with TensorFlow Lite for Microcontrollers and the Arduino Nano
Technical requirements
Taking pictures with the OV7670 camera module
Grabbing camera frames from the serial port with Python
Converting QQVGA images from YCbCr422 to RGB888
Building the dataset for indoor scene classification
Transfer learning with Keras
Preparing and testing the quantized TFLite model
Reducing RAM usage by fusing crop, resize, rescale, and quantize
6
Chapter 6: Building a Gesture-Based Interface for YouTube Playback
Chapter 6: Building a Gesture-Based Interface for YouTube Playback
Technical requirements
Communicating with the MPU-6050 IMU through I2C
Acquiring accelerometer data
Building the dataset with the Edge Impulse data forwarder tool
Designing and training the ML model
Live classifications with the Edge Impulse data forwarder tool
Gesture recognition on Raspberry Pi Pico with Arm Mbed OS
Building a gesture-based interface with PyAutoGUI
7
Chapter 7: Running a Tiny CIFAR-10 Model on a Virtual Platform with the Zephyr OS
Chapter 7: Running a Tiny CIFAR-10 Model on a Virtual Platform with the Zephyr OS
Technical requirements
Getting started with the Zephyr OS
Designing and training a tiny CIFAR-10 model
Evaluating the accuracy of the TFLite model
Converting a NumPy image to a C-byte array
Preparing the skeleton of the TFLu project
Building and running the TFLu application on QEMU
Join us on Discord!
8
Chapter 8: Toward the Next TinyML Generation with microNPU
Chapter 8: Toward the Next TinyML Generation with microNPU
Technical requirements
Setting up Arm Corstone-300 FVP
Installing TVM with Arm Ethos-U support
Installing the Arm toolchain and Ethos-U driver stack
Generating C code with TVM
Generating C-byte arrays for input, output, and labels
Building and running the model on Ethos-U55
Why subscribe?
9
Other Books You May Enjoy
Other Books You May Enjoy
Packt is searching for authors like you
Share Your Thoughts

To get the most out of this book

You will need a computer (either a laptop or desktop) with an x86-64 architecture and at least one USB port for programming the Arduino Nano 33 BLE Sense and the Raspberry Pi Pico microcontroller boards. For the first six chapters, you can use Ubuntu 18.04 (or later) or Windows (for example, Windows 10) as an OS. However, you will need Ubuntu 18.04 (or later) for Chapter 7, Running a Tiny CIFAR-10 Model on a Virtual Platform with the Zephyr OS, and Chapter 8, Toward the Next TinyML Generation with microNPU.

The only software prerequisites for your computer are:

  • Python (Python 3.7 recommended)
  • Text editor (for example, gedit on Ubuntu)
  • Media player (for example, VLC)
  • Image viewer (for example, the default app in Ubuntu or Windows 10)
  • Web browser (for example, Google Chrome)

During our TinyML journey, we will require different software tools to cover ML development and embedded programming. Thanks to Arduino, Edge Impulse, and Google, these tools will be in the cloud, browser-based, and with a free plan for our usage.

Arduino Nano 33 BLE Sense and Raspberry Pi Pico programs will be developed directly in the web browser with the Arduino Web Editor (https://create.arduino.cc). However, the Arduino Web Editor has a limit of 200 seconds of compilation time per day. Therefore, you may consider upgrading to any paid plan or using the free local Arduino IDE (https://www.arduino.cc/en/software) to get unlimited compilation time. If you are interested in the free local Arduino IDE, we have provided on GitHub (https://github.com/PacktPublishing/TinyML-Cookbook/blob/main/Docs/setup_local_arduino_ide.md) the instructions to set it up.

The following table summarizes the hardware devices and software tools covered in each chapter:

The projects may require sensors and additional electronic components to build realistic TinyML prototypes and experience the complete development workflow. All the components are listed at the beginning of each chapter and in the README.md file on GitHub (https://github.com/PacktPublishing/TinyML-Cookbook). Since you will build real electronic circuits, we require an electronic components kit that includes at least a solderless breadboard, colored LEDs, resistors, push-buttons, and jumper wires. Don't worry if you are an electronics beginner. You will learn more about these components in the first two chapters of this book. Furthermore, we have prepared a beginner shopping list on GitHub so you know precisely what to buy: https://github.com/PacktPublishing/TinyML-Cookbook/blob/main/Docs/shopping_list.md.

If you are using the digital version of this book, we advise you to type the code yourself or access the code via the GitHub repository (link available in the next section). Doing so will help you avoid any potential errors related to the copying and pasting of code.

Download the example code files

You can download the example code files for this book from GitHub at https://github.com/PacktPublishing/TinyML-Cookbook. In case there's an update to the code, it will be updated on the existing GitHub repository.

We also have other code bundles from our rich catalog of books and videos available at https://github.com/PacktPublishing/. Check them out!

Previous Section
Next Section