AI and Machine Learning for On-Device Development: A Programmer's Guide

Cover
Copyright
Table of Contents
Preface
	Who Should Read This Book?
	Why I Wrote This Book
	Navigating This Book
	Technology You Need to Understand
	Conventions Used in This Book
	Using Code Examples
	O’Reilly Online Learning
	How to Contact Us
	Acknowledgements
Chapter 1. Introduction to AI and Machine Learning
	What Is Artificial Intelligence?
	What Is Machine Learning?
		Moving from Traditional Programming to Machine Learning
		How Can a Machine Learn?
		Comparing Machine Learning with Traditional Programming
	Building and Using Models on Mobile
	Summary
Chapter 2. Introduction to Computer Vision
	Using Neurons for Vision
		Your First Classifier: Recognizing Clothing Items
		The Data: Fashion MNIST
		A Model Architecture to Parse Fashion MNIST
		Coding the Fashion MNIST Model
	Transfer Learning for Computer Vision
	Summary
Chapter 3. Introduction to ML Kit
	Building a Face Detection App on Android
		Step 1: Create the App with Android Studio
		Step 2: Add and Configure ML Kit
		Step 3: Define the User Interface
		Step 4: Add the Images as Assets
		Step 5: Load the UI with a Default Picture
		Step 6: Call the Face Detector
		Step 7: Add the Bounding Rectangles
	Building a Face Detector App for iOS
		Step 1: Create the Project in Xcode
		Step 2: Using CocoaPods and Podfiles
		Step 3: Create the User Interface
		Step 4: Add the Application Logic
	Summary
Chapter 4. Computer Vision Apps with ML Kiton Android
	Image Labeling and Classification
		Step 1: Create the App and Configure ML Kit
		Step 2: Create the User Interface
		Step 3: Add the Images as Assets
		Step 4: Load an Image to the ImageView
		Step 5: Write the Button Handler Code
		Next Steps
	Object Detection
		Step 1: Create the App and Import ML Kit
		Step 2: Create the Activity Layout XML
		Step 3: Load an Image into the ImageView
		Step 4: Set Up the Object Detector Options
		Step 5: Handling the Button Interaction
		Step 6: Draw the Bounding Boxes
		Step 7: Label the Objects
	Detecting and Tracking Objects in Video
		Exploring the Layout
		The GraphicOverlay Class
		Capturing the Camera
		The ObjectAnalyzer Class
		The ObjectGraphic Class
		Putting It All Together
	Summary
Chapter 5. Text Processing Apps with ML Kiton Android
	Entity Extraction
		Start Creating the App
		Create the Layout for the Activity
		Write the Entity Extraction Code
		Putting It All Together
	Handwriting and Other Recognition
		Start the App
		Creating a Drawing Surface
		Parsing the Ink with ML Kit
	Smart Reply to Conversations
		Start the App
		Mock a Conversation
		Generating a Smart Reply
	Summary
Chapter 6. Computer Vision Apps with ML Kit on iOS
	Image Labeling and Classification
		Step 1: Create the App in Xcode
		Step 2: Create the Podfile
		Step 3: Set Up the Storyboard
		Step 4: Edit the View Controller Code to Use ML Kit
	Object Detection in iOS with ML Kit
		Step 1: Get Started
		Step 2: Create Your UI on the Storyboard
		Step 3: Create a Subview for Annotation
		Step 4: Perform the Object Detection
		Step 5: Handle the Callback
		Combining Object Detection with Image Classification
		Object Detection and Tracking in Video
	Summary
Chapter 7. Text Processing Apps with ML Kit on iOS
	Entity Extraction
		Step 1: Create the App and Add the ML Kit Pods
		Step 2: Create the Storyboard with Actions and Outlets
		Step 3: Allow Your View Controller to be Used for Text Entry
		Step 4: Initialize the Model
		Step 5: Extract Entities from Text
	Handwriting Recognition
		Step 1: Create the App and Add the ML Kit Pods
		Step 2: Create the Storyboard, Actions, and Outlets
		Step 3: Strokes, Points, and Ink
		Step 4: Capture User Input
		Step 5: Initialize the Model
		Step 6: Do the Ink Recognition
	Smart Reply to Conversations
		Step 1: Create an App and Integrate ML Kit
		Step 2: Create Storyboard, Outlets, and Actions
		Step 3: Create a Simulated Conversation
		Step 4: Get Smart Reply
	Summary
Chapter 8. Going Deeper: Understanding  TensorFlow Lite
	What Is TensorFlow Lite?
	Getting Started with TensorFlow Lite
		Save the Model
		Convert the Model
		Testing the Model with a Standalone Interpreter
	Create an Android App to Host TFLite
		Import the TFLite File
		Write Kotlin Code to Interface with the Model
		Going Beyond the Basics
	Create an iOS App to Host TFLite
		Step 1: Create a Basic iOS App
		Step 2: Add TensorFlow Lite to Your Project
		Step 3: Create the User Interface
		Step 4: Add and Initialize the Model Inference Class
		Step 5: Perform the Inference
		Step 6: Add the Model to Your App
		Step 7: Add the UI Logic
		Moving Beyond “Hello World”: Processing Images
	Exploring Model Optimization
		Quantization
		Using Representative Data
	Summary
Chapter 9. Creating Custom Models
	Creating a Model with TensorFlow Lite Model Maker
	Creating a Model with Cloud AutoML
		Using AutoML Vision Edge
	Creating a Model with TensorFlow and Transfer Learning
	Creating Language Models
		Create a Language Model with Model Maker
	Summary
Chapter 10. Using Custom Models in Android
	Bridging Models to Android
	Building an Image Classification App from a Model Maker Output
	Using a Model Maker Output with ML Kit
	Using Language Models
	Creating an Android App for Language Classification
	Summary
Chapter 11. Using Custom Models in iOS
	Bridging Models to iOS
	A Custom Model Image Classifier
		Step 1: Create the App and Add the TensorFlow Lite Pod
		Step 2: Create the UI and Image Assets
		Step 3: Load and Navigate Through the Image Assets
		Step 4: Load the Model
		Step 5: Convert an Image to an Input Tensor
		Step 6: Get Inference for the Tensor
	Use a Custom Model in ML Kit
	Building an App for Natural Language Processing in Swift
		Step 1: Load the Vocab
		Step 2: Convert the Sentence to a Sequence
		Step 3: Extend Array to Handle Unsafe Data
		Step 4: Copy the Array to a Data Buffer
		Step 5: Run Inference on the Data and Process the Results
	Summary
Chapter 12. Productizing Your App Using Firebase
	Why Use Firebase Custom Model Hosting?
	Create Multiple Model Versions
	Using Firebase Model Hosting
		Step 1: Create a Firebase Project
		Step 2: Use Custom Model Hosting
		Step 3: Create a Basic Android App
		Step 4: Add Firebase to the App
		Step 5: Get the Model from Firebase Model Hosting
		Step 6: Use Remote Configuration
		Step 7: Read Remote Configuration in Your App
		Next Steps
	Summary
Chapter 13. Create ML and Core ML for Simple iOS Apps
	A Core ML Image Classifier Built Using Create ML
		Making a Core ML App That Uses a Create ML Model
		Add the MLModel File
		Run the Inference
	Using Create ML to Build a Text Classifier
	Use the Model in an App
	Summary
Chapter 14. Accessing Cloud-Based Models from Mobile Apps
	Installing TensorFlow Serving
		Installing Using Docker
		Installing Directly on Linux
	Building and Serving a Model
	Accessing a Server Model from Android
	Accessing a Server Model from iOS
	Summary
Chapter 15. Ethics, Fairness, and Privacy for Mobile Apps
	Ethics, Fairness, and Privacy with Responsible AI
		Responsibly Defining Your Problem
		Avoiding Bias in Your Data
		Building and Training Your Model
		Evaluating Your Model
	Google’s AI Principles
	Summary
Index
About the Author
Colophon