Do Machine Learning Yourself with Python
Do Machine Learning Yourself with Python
About the Book
Do Machine Learning Yourself is a collection of do it yourself (DIY) projects about machine learning, mainly about computer vision, for beginner and intermediate levels. Through a detailed guidance per each project, everything required to do that project yourself will be clear. One primary focus of the book is to make machine learning available for mobile devices.
The book is organized into 8 chapters where each chapter handles a topic.
Chapter 1 is about doing some machine learning projects using OpenCV. Because OpenCV is a cross-platform library, then the projects can run on both desktop and mobile devices. For such a reason, the book builds the project for running in Android devices.
Chapter 2 builds 3 computer vision projects for Raspberry Pi. This includes preparing Raspberry Pi for first-time use in order to be able to access it and install some Python libraries. A USB camera is connected to the Raspberry Pi to capture images for building a surveillance system and also a simple robot car.
Chapter 3 discusses some topics about neural network such as selecting the best architecture that works for a given data. Also, a neural network is created in Keras that is deployed to a Flask server. An Android client uploads images to be classified by the trained Keras model and then return the response.
Chapter 4 uses the genetic algorithm for building doing some optimization. This includes optimizing the 8 queen puzzle to reach a solution. Besides that project, the algorithm is used for reproducing images by evolving the pixel values.
Chapter 5 uses a framework named Kivy for building GUI Python applications. Based on the Kivy app, cross-platform applications can be generated. This book focuses on building Android applications with the help of 2 projects named Python-4-Android and Buildozer. Kivy allows some Python libraries to be executed in Android such as NumPy which is one of the most important libraries for building data science applications. An image classifier is created in Python that is able to work in Android using NumPy. The book also discusses supporting the Arabic text with Kivy.
Chapter 6 uses OpenCV for building an Android application able to do some image effects. The effects supported are horizontal and vertical image stitching, reducing the number of colors representing the both color and gray images by editing the lookup tables, noise removal using the median filter, converting images to binary, cartooning images, image blending using the alpha channel, creating animated GIF image from a number of images, and creating image patterns.
Chapter 7 builds a chat application for Android devices for sending and receiving text messages. The app allows users to register themselves by entering their information that includes a username, password, and e-mail. One the e-mail address is verified, the user is registered and able to send and receive messages. The project starts by building the database and the tables for holding the users' data and also the messages. A Flask-based server is created that is connected to the Android app. The project supports instant notifications for new messages and encryption.
Chapter 8 lists some miscellaneous topics about machine learning. This includes a discussion about how machine learning is not killed by the appearance of deep learning. Also, this chapter guides beginners to understand how ensemble algorithms work and understand the difference between bagging and boosting ensemble models. The gradient boosting algorithm is discussed which is an application of ensemble boosting.
About the Author
Table of Contents
Table of Contents
Preface
Acknowledgment
About the Author
Table of Contents
Chapter 1: Machine Learning with OpenCV
A Guide to Preparing OpenCV for Android
Overview of OpenCV
Building an Android Studio Project
Running the Project
Editing the Project to Display a Toast Message
Downloading OpenCV
Importing OpenCV in Android Studio
Fixing Possible Gradle Sync Errors
Adding OpenCV as a Dependency
Adding Native Libraries
Using OpenCV for Filtering Images
Summary
Running Artificial Neural Networks in Android using OpenCV
Creating a New Java Application in NetBeans
Downloading OpenCV for Windows
Importing OpenCV in NetBeans
Solving the UnsatisfiedLinkError
Preparing the Training Data
Building the ANN (Architecture and Parameters)
Training the ANN
Loading and Testing the Trained ANN in NetBeans
Building an Android Studio Project
Loading the Trained ANN in Android Studio and Making Predictions
Conclusion
Image Classification on Android using OpenCV
Preparing the Image Dataset
Image Feature Extraction (Color Histogram)
Building, Training, and Saving the ANN
Making Predictions by Loading the Saved ANN
Building an Android App
Predicting the Class Label of the Image in Android
Project at GitHub
Chapter 2: Machine Learning for Raspberry Pi
Building an Image Classifier Running on Raspberry Pi
Raspberry Pi
Network Configuration
Secure Shell
Login
X11 Windowing System
Image Classification
For More Details
Building Surveillance System using USB Camera and Wireless-Connected Raspberry Pi
Connecting RPi to a PC using the Wireless Interface
Connecting a USB Camera to RPi
Capturing Images using PyGame
Building the Background Model
Detecting Changes to the Background Model
Building a Simple Circuit that Lights a Led When a Change Occurs
For More Details
Building a Vision-Controlled Car Using Raspberry Pi—From Scratch
Preparing the Car Components
Fixing the Components on the Wood Plank
Understanding the Circuit Connected to the Motor
Building the Circuit on the BB using Transistors, Resistors, and Diodes
Connecting the BB to the Raspberry Pi GPIO Pins
Accessing the Raspberry Pi
Controlling the Car’s Motors
Connecting a USB Camera to the Raspberry Pi
Controlling the Motors using the Camera
For More Details
Chapter 3: Neural Networks
Selecting the Best Architecture for Artificial Neural Networks
Example 1
Python Implementation
Example 2
Conclusion
Uploading images from Android to a Python-based Flask server
Building the Layout of the Android App
Using OkHttp for building the client-side Android app
Building the server-side Python app using Flask
Testing the connection between Android and Flask
Selecting an image from Android storage
Modifying the Flask app to save an uploaded image
Conclusion
Image Classification on Android using a Keras Model Deployed in Flask
Preparing the MNIST Dataset
Building the MLP
Training the MLP
Saving the Model
Loading the Model and Making Predictions
Editing the Flask Server to Classify Uploaded Images
Conclusion
Chapter 4: Optimization with Genetic Algorithm
8 Queen Puzzle Optimization Using a Genetic Algorithm in Python
8 Queen Puzzle Review
Building the 8x8 Board using Kivy
Adding Widgets to Control the Project
Initializing the Population
Visualizing the Best Solution with a Population
Optimization using GA
Running the Project
Reproducing Images using Genetic Algorithm with Python
Further Reading about GA and its Python Implementation
Genetic Algorithm Steps
Data Representation
Python Code for Converting the Image into a Chromosome and Vice Versa
Initial Population
Fitness Calculation
Parent Selection
Crossover
Mutation
Project Example
Conclusion + 2 Helper Functions
Chapter 5: Python for Android with Kivy
Python for Android: Start Building Kivy Cross-Platform Applications
Introduction
Installing Kivy Dependencies
Creating Virtual Environment for Installing Kivy
Installing Cython
Installing Kivy
Importing Kivy
Creating Simple Kivy Application
Installing Buildozer
Creating buildozer.spec File
Building Android Application using Buildozer
References
Running NumPy in Android Devices using the Kivy Python Framework
Introduction
Kivy Installation for Linux
Installing Kivy in a Virtual Environment
Importing Kivy
Creating a Basic Application
Python for Android
Installing Buildozer
Preparing buildozer.spec File
Building an Android Application using Buildozer
Image Classification for Android Devices Using NumPy and Kivy
For More Information
ANN Architecture
Creating the Widget Tree using the KV Language
Creating the Kivy Application
Using the Proper NumPy Version
Building the Android Application
Supporting Arabic Alphabet in Kivy for Building Cross-Platform Applications
Introduction
Displaying Arabic Text on a Label Widget
Using a Font that Supports the Arabic Alphabet
Changing the Text Direction using BIDI Algorithm
Reshaping the Text using the Python Arabic Reshaper Library
Editing the TextInput Class
FileChooser
Conclusion
Chapter 6: Image Effects using OpenCV for Android
Vertical and Horizontal Stitching
Preparing OpenCV for Android
Image Stitching
Horizontal Image Stitching
Horizontal Stitching Implementation
Saving Images to the Gallery
Uploading Images from Android to a Flask Server
Vertical Image Stitching
Conclusion
Cartoon Effect
What is LUT?
OpenCV Built-in LUTs
Color Reduction using LUT for Gray Images
Color Reduction Using LUT for Color Images
Using a Single LUT for All Color Image Channels
Using a Single LUT for Each Color Channel
Adaptive Thresholding
Noise Removal Using Median Filter
Cartoon Effect
Building Android App
Conclusion
Image Blending
What is Alpha Channel?
Working with Alpha Channel in OpenCV
Image Blending
Region Blending
Building the Android App
Conclusion
Animated GIF
Using GridLayout
Selecting a Single Image
Selecting More than One Image
Applying Effects Using Selected Images
Reading Multiple Images for Multi-Image Effects
Create Animated GIF
Building Android App
Conclusion
Image Pattern
A Review about 2D Convolution
Creating Patterns
Adding a Pattern to an Image
Editing Android Studio Project
Conclusion
Chapter 7: Chat Application for Android
Android Login System
Creating the MySQL Database
Connecting to MySQL
Returning a Cursor
Executing Operations
Creating a Table for the Users Data
Building the Flask Server
Building the Android App Main Screen
Registration Activity
Login Activity
Testing the System
Conclusion
E-mail Verification for an Android Registration System
Sending E-mails using Python
Verification Code
Text Encryption
Generating Encryption Keys
Verifying Code
Text Decryption
Adding an EditText for E-mail in the Android App
Conclusion
Sending Text Messages for Android between Verified Users
Chapter 8: Others
Do New Technologies Kill their Ancestors?
Intuitive Ensemble Learning Guide with Gradient Boosting as a Study Case
Introduction
Ensemble Learning
Gradient Boosting (GB)
Summary of GB
The Leanpub 60 Day 100% Happiness Guarantee
Within 60 days of purchase you can get a 100% refund on any Leanpub purchase, in two clicks.
Now, this is technically risky for us, since you'll have the book or course files either way. But we're so confident in our products and services, and in our authors and readers, that we're happy to offer a full money back guarantee for everything we sell.
You can only find out how good something is by trying it, and because of our 100% money back guarantee there's literally no risk to do so!
So, there's no reason not to click the Add to Cart button, is there?
See full terms...
Earn $8 on a $10 Purchase, and $16 on a $20 Purchase
We pay 80% royalties on purchases of $7.99 or more, and 80% royalties minus a 50 cent flat fee on purchases between $0.99 and $7.98. You earn $8 on a $10 sale, and $16 on a $20 sale. So, if we sell 5000 non-refunded copies of your book for $20, you'll earn $80,000.
(Yes, some authors have already earned much more than that on Leanpub.)
In fact, authors have earnedover $14 millionwriting, publishing and selling on Leanpub.
Learn more about writing on Leanpub
Free Updates. DRM Free.
If you buy a Leanpub book, you get free updates for as long as the author updates the book! Many authors use Leanpub to publish their books in-progress, while they are writing them. All readers get free updates, regardless of when they bought the book or how much they paid (including free).
Most Leanpub books are available in PDF (for computers) and EPUB (for phones, tablets and Kindle). The formats that a book includes are shown at the top right corner of this page.
Finally, Leanpub books don't have any DRM copy-protection nonsense, so you can easily read them on any supported device.
Learn more about Leanpub's ebook formats and where to read them
