Machine Learning for Mobile电子书

Title Page

Copyright and Credits

Machine Learning for Mobile

About Packt

Why subscribe?

Packt.com

Contributors

About the authors

About the reviewer

Packt is searching for authors like you

Preface

Who this book is for

What this book covers

To get the most out of this book

Download the example code files

Download the color images

Conventions used

Get in touch

Reviews

Introduction to Machine Learning on Mobile

Definition of machine learning

When is it appropriate to go for machine learning systems?

The machine learning process

Defining the machine learning problem

Preparing the data

Building the model

Selecting the right machine learning algorithm

Training the machine learning model

Testing the model

Evaluation of the model

Making predictions/Deploying in the field

Types of learning

Supervised learning

Unsupervised learning

Semi-supervised learning

Reinforcement learning

Challenges in machine learning

Why use machine learning on mobile devices?

Ways to implement machine learning in mobile applications

Utilizing machine learning service providers for a machine learning model

Ways to train the machine learning model

On a desktop (training in the cloud)

On a device

Ways to carry out the inference – making predictions

Inference on a server

Inference on a device

Popular mobile machine learning tools and SDKs

Skills needed to implement on-device machine learning

Summary

Supervised and Unsupervised Learning Algorithms

Introduction to supervised learning algorithms

Deep dive into supervised learning algorithms

Naive Bayes

Decision trees

Linear regression

Logistic regression

Support vector machines

Random forest

Introduction to unsupervised learning algorithms

Deep dive into unsupervised learning algorithms

Clustering algorithms

Clustering methods

Hierarchical agglomerative clustering methods

K-means clustering

Association rule learning algorithm

Summary

References

Random Forest on iOS

Introduction to algorithms

Decision tree

Advantages of the decision tree algorithm

Disadvantages of decision trees

Advantages of decision trees

Random forests

Solving the problem using random forest in Core ML

Dataset

Naming the dataset

Technical requirements

Creating the model file using scikit-learn

Converting the scikit model to the Core ML model

Creating an iOS mobile application using the Core ML model

Summary

Further reading

TensorFlow Mobile in Android

An introduction to TensorFlow

TensorFlow Lite components

Model-file format

Interpreter

Ops/Kernel

Interface to hardware acceleration

The architecture of a mobile machine learning application

Understanding the model concepts

Writing the mobile application using the TensorFlow model

Writing our first program

Creating and Saving the TF model

Freezing the graph

Optimizing the model file

Creating the Android app

Copying the TF Model

Creating an activity

Summary

Regression Using Core ML in iOS

Introduction to regression

Linear regression

Dataset

Dataset naming

Understanding the basics of Core ML

Solving the problem using regression in Core ML

Technical requirements

How to create the model file using scikit-learn

Running and testing the model

Importing the model into the iOS project

Writing the iOS application

Running the iOS application

Further reading

Summary

The ML Kit SDK

Understanding ML Kit

ML Kit APIs

Text recognition

Face detection

Barcode scanning

Image labeling

Landmark recognition

Custom model inference

Creating a text recognition app using Firebase on-device APIs

Creating a text recognition app using Firebase on-cloud APIs

Face detection using ML Kit

Face detection concepts

Sample solution for face detection using ML Kit

Running the app

Summary

Spam Message Detection

Understanding NLP

Introducing NLP

Text-preprocessing techniques

Removing noise

Normalization

Standardization

Feature engineering

Entity extraction

Topic modeling

Bag-of-words model

Statistical Engineering

TF–IDF

TF

Inverse Document Frequency (IDF)

TF-IDF

Classifying/clustering the text

Understanding linear SVM algorithm

Solving the problem using linear SVM in Core ML

About the data

Technical requirements

Creating the Model file using Scikit Learn

Converting the scikit-learn model into the Core ML model

Writing the iOS application

Summary

Fritz

Introduction to Fritz

Prebuilt ML models

Ability to use custom models

Model management

Hand-on samples using Fritz

Using the existing TensorFlow for mobile model in an Android application using Fritz

Registering with Fritz

Uploading the model file (.pb or .tflite)

Setting up Android and registering the app

Adding Fritz's TFMobile library

Adding dependencies to the project

Registering the FritzJob service in your Android Manifest

Replacing the TensorFlowInferenceInterface class with Fritz Interpreter

Building and running the application

Deploying a new version of your model

Creating an android application using fritz pre-built models

Adding dependencies to the project

Registering the Fritz JobService in your Android Manifest

Creating the app layout and components

Coding the application

Using the existing Core ML model in an iOS application using Fritz

Registering with Fritz

Creating a new project in Fritz

Uploading the model file (.pb or .tflite)

Creating an Xcode project

Installing Fritz dependencies

Adding code

Building and running the iOS mobile application

Summary

Neural Networks on Mobile

Introduction to neural networks

Communication steps of a neuron

The activation function

Arrangement of neurons

Types of neural networks

Image recognition solution

Creating a TensorFlow image recognition model

What does TensorFlow do?

Retraining the model

About bottlenecks

Converting the TensorFlow model into the Core ML model

Writing the iOS mobile application

Handwritten digit recognition solution

Introduction to Keras

Installing Keras

Solving the problem

Defining the problem statement

Problem solution

Preparing the data

Defining the model's architecture

Compiling and fitting the model

Converting the Keras model into the Core ML model

Creating the iOS mobile application

Summary

Mobile Application Using Google Vision

Features of Google Cloud Vision

Sample mobile application using Google Cloud Vision

How does label detection work?

Prerequisites

Preparations

Understanding the Application

Output

Summary

The Future of ML on Mobile Applications

Key ML mobile applications

Facebook

Google Maps

Snapchat

Tinder

Netflix

Oval Money

ImprompDo

Dango

Carat

Uber

GBoard

Key innovation areas

Personalization applications

Healthcare

Targeted promotions and marketing

Visual and audio recognition

E-commerce

Finance management

Gaming and entertainment

Enterprise apps

Real estate

Agriculture

Energy

Mobile security

Opportunities for stakeholders

Hardware manufacturers

Mobile operating system vendors

Third-party mobile ML SDK providers

ML mobile application developers

Summary

Question and Answers

FAQs

Data science

What is data science?

Where is data science used?

What is big data?

What is data mining?

Relationship between data science and big data

What are artificial neural networks?

What is AI?

How are data science, AI, and machine learning interrelated?

Machine learning framework

Caffe2

scikit-learn

TensorFlow

Core ML

Mobile machine learning project implementation

What are the high-level important items to be considered before starting the project?

What are the roles and skills required to implement a mobile machine learning project?

What should you focus on when testing the mobile machine learning project?

What is the help that the domain expert will provide to the machine learning project?

What are the common pitfalls in machine learning projects?

Installation

Python

Python dependencies

Xcode

References

Other Books You May Enjoy

Leave a review - let other readers know what you think