book

TinyML Cookbook

Name: TinyML Cookbook
Author: Gian Marco Iodice
ISBN: 9781801814973

by Gian Marco Iodice

April 2022

Intermediate to advanced

344 pages

7h 18m

English

Packt Publishing

Read now

Unlock full access

TinyML Cookbook
ForewordContributorsAbout the authorAbout the reviewers
Preface
Who this book is forWhat this book coversTo get the most out of this bookDownload the example code filesDownload the color imagesConventions usedSectionsGetting readyHow to do it…There's more…Get in touchShare Your Thoughts
Chapter 1: Getting Started with TinyML
Technical requirementsIntroducing TinyMLWhat is TinyML?Why ML on microcontrollers?Why run ML locally?The opportunities and challenges for TinyMLDeployment environments for TinyMLtinyML FoundationSummary of DLDeep neural networksConvolutional neural networksQuantizationLearning the difference between power and energyVoltage versus currentPower versus energyProgramming microcontrollersMemory architecturePeripheralsPresenting Arduino Nano 33 BLE Sense and Raspberry Pi PicoSetting up Arduino Web Editor, TensorFlow, and Edge ImpulseGetting ready with Arduino Web EditorGetting ready with TensorFlowGetting ready with Edge ImpulseHow to do it…Running a sketch on Arduino Nano and Raspberry Pi PicoGetting readyHow to do it…Join us on Discord!
Chapter 2: Prototyping with Microcontrollers
Technical requirementsCode debugging 101Getting readyHow to do it...There's moreImplementing an LED status indicator on the breadboardGetting readyHow to do it...Controlling an external LED with the GPIOGetting readyHow to do it...Turning an LED on and off with a push-buttonGetting readyHow to do it...Using interrupts to read the push-button stateGetting readyHow to do it...Powering microcontrollers with batteriesGetting startedHow to do it...There's more
Chapter 3: Building a Weather Station with TensorFlow Lite for Microcontrollers
Technical requirementsImporting weather data from WorldWeatherOnlineGetting readyHow to do it…Preparing the datasetGetting readyHow to do it…Training the ML model with TFGetting readyHow to do it…Evaluating the model's effectivenessGetting readyHow to do it…Quantizing the model with the TFLite converterGetting readyHow to do it…Using the built-in temperature and humidity sensor on Arduino NanoGetting readyHow to do it…Using the DHT22 sensor with the Raspberry Pi PicoGetting readyHow to do it…Preparing the input features for the model inferenceGetting readyHow to do it…On-device inference with TFLuGetting readyHow to do it…
Chapter 4: Voice Controlling LEDs with Edge Impulse
Technical requirementsAcquiring audio data with a smartphoneGetting readyHow to do it…Extracting MFCC features from audio samplesGetting readyHow to do it…There's more…Designing and training a NN modelGetting readyHow to do it…Tuning model performance with EON TunerGetting readyHow to do it…Live classifications with a smartphoneGetting readyHow to do it…Live classifications with the Arduino NanoGetting readyHow to do it…Continuous inferencing on the Arduino NanoGetting readyHow to do it…Building the circuit with the Raspberry Pi Pico to voice control LEDsGetting readyHow to do it…Audio sampling with ADC and timer interrupts on the Raspberry Pi PicoGetting readyHow to do it…There's more…
Chapter 5: Indoor Scene Classification with TensorFlow Lite for Microcontrollers and the Arduino Nano
Technical requirementsTaking pictures with the OV7670 camera moduleGetting readyHow to do it...Grabbing camera frames from the serial port with PythonGetting readyHow to do it...Converting QQVGA images from YCbCr422 to RGB888Getting readyHow to do it...Building the dataset for indoor scene classificationGetting readyHow to do it...Transfer learning with KerasGetting readyHow to do it...Preparing and testing the quantized TFLite modelGetting readyHow to do it...Reducing RAM usage by fusing crop, resize, rescale, and quantizeGetting readyHow to do it...
Chapter 6: Building a Gesture-Based Interface for YouTube Playback
Technical requirementsCommunicating with the MPU-6050 IMU through I2CGetting readyHow to do it…Acquiring accelerometer dataGetting readyHow to do it…Building the dataset with the Edge Impulse data forwarder toolGetting readyHow to do it…Designing and training the ML modelGetting readyHow to do it…Live classifications with the Edge Impulse data forwarder toolGetting readyHow to do it…Gesture recognition on Raspberry Pi Pico with Arm Mbed OSGetting readyHow to do it…Building a gesture-based interface with PyAutoGUIGetting readyHow to do it…
Chapter 7: Running a Tiny CIFAR-10 Model on a Virtual Platform with the Zephyr OS
Technical requirementsGetting started with the Zephyr OSGetting readyHow to do it…Designing and training a tiny CIFAR-10 modelGetting readyHow to do it…Evaluating the accuracy of the TFLite modelGetting readyHow to do it…Converting a NumPy image to a C-byte array Getting readyHow to do it…Preparing the skeleton of the TFLu projectGetting readyHow to do it…Building and running the TFLu application on QEMUGetting readyHow to do it…Join us on Discord!
Chapter 8: Toward the Next TinyML Generation with microNPU
Technical requirementsSetting up Arm Corstone-300 FVPGetting readyHow to do it…Installing TVM with Arm Ethos-U supportGetting readyHow to do it…Installing the Arm toolchain and Ethos-U driver stackGetting readyHow to do it…Generating C code with TVMGetting readyHow to do it…Generating C-byte arrays for input, output, and labelsGetting readyHow to do it…Building and running the model on Ethos-U55Getting readyHow to do it…

Why subscribe?
Other Books You May EnjoyPackt is searching for authors like youShare Your Thoughts

Content preview from TinyML Cookbook

Chapter 8: Toward the Next TinyML Generation with microNPU

Here, we are at the last stop of our journey into the world of TinyML. Although this chapter may look like the end, it is actually the beginning of something new and extraordinary for Machine Learning (ML) at the very edge. In our journey, we have learned how vital power consumption is for effective and long-lasting TinyML applications. However, computing capacity is the key to unlocking new use cases and making the "things" around us even more intelligent. For this reason, a new, advanced processor has been designed to extend the computational power and energy efficiency of ML workloads. This processor is the Micro-Neural Processing Unit (microNPU).

In this final chapter, we will discover ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Publisher Resources

ISBN: 9781801814973

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

TinyML Cookbook

by Gian Marco Iodice

Chapter 8: Toward the Next TinyML Generation with microNPU

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.