Practical Deep Learning Fundamentals

This book is designed to provide a self-teaching platform to Machine Learning, with the focus on Deep Learning. You will learn concept through to implementation using open source libaries - Google’s TensorFlow and Python’s Scikit-learn. This book aims to be different to other texts by being designed to reduce the learning curve associated with Machine Learning from the outset.

“designed to reduce the learning curve” ― Kris Bolton ―


Chapter One: Why You Should get into Machine Learning
A discussion of the mergers, aquisitions, commercial demand, breakthroughts and more that make machine learning a valuable and intereting area to add to your resume.
Chapter Two: An Introduction to Machine Learning
An introduction to the broad area of machine learning. It's helpful to understand where we've come from to see where we're heading.
Chapter Three: An Overview of Open Source Machine Learning Libraries
An overview introduction to the most popular machine learning libraries, giving the reader a broad understanding of the open source landscape for their own projects.
Chapter Four: A Breif Tour of Non-Deep Methods in Machine Learning
A short chapter on a number of popular methods in machine learning which are not deep neural networks. One aim of the book is to leave readers with a greater broad understanding of the area so they can move beyond what is within the book themselves.
Chapter Six: An Introduction to Artificial Neural Networks
Artificial Neural Networks (ANNs) and the concepts surrounding them are introduced and illustrated. The reader is presented with a practical tas to implement a single neuron program and analyse a single data set.
Chapter Seven: How to Train Artificial Neural Networks
How neural networks learn and the various pit-falls of techniques are discussed. As the reader continues through the book techniques which have been used are explained further and are used within practical examples.
Chapter Eight: Gathering Data and Creating Datasets
Methods for gathering data and creating datasets, consdiering how the data is to be used and how the neural network will use this data is discussed within chapter eight.
Chapter Nine: An Introduction to TensorFlow
TensorFlow is Google's machine learning libary. It is by far the most popular machine learning library in the world. Additionally, preceeding capters use TensorFlow for practical examples, the introduction allows the reader to complete future exercises.
Chapter Ten: An Introduction to Deep Learning
A breif introduction to deep learning is provided and the areas which deep learning provides significat advantages is described.
Chapter Eleven: Foundational Algorithms in Deep Learning
The foundational algorithms of deep learning are discussed. Perceptron, Markov Chain, Hopfield Networks and Boltzmann Machines are described.
Chapter Tweleve: Algorithms in Deep Learning
Foundational algorithms are the historic algorithms on which most modern algorithms are built or inspired from, algorithms discussed within this chapter are modren state-of-the-art.
Chapter Thirteen: How to Evaluate and Tune Neural Networks
Evaluating neural network performance and outcomes, and tuning the network to improve aspects are learned skills. This chapter aims to teach the reader how to perform these actions to get desired outcomes and ensure performance.
Chapter Fourteen: Project: Sentiment Analysis
Sentiment Analysis, deriving feeling from text is an important technique which as wide applicaiton; from intelligence gathering to marketing. This chapter aims to provide an introduction to the area and techniques via a practical project.
Chapter Fifteen: Project: Object Recognition
Object recognition is a growing and important area within robotics, sensors and has implementations likely not thought of yet. The reader is introduced to the area and guided through examples to learn the techniques for themselves.
Chapter Sixteen: Project: Reinforcement Learning
Reinforcement learning, the process of a machine learning from its own mistakes is a powerful technique which allows machines to learn on their own, creating outcomes far surpassing what humans can achieve. The reader is guided through techniques in a practical format.
Chapter Seventeen: Understanding Neural Network Math
A small amount of math is used throughout this book, however, some readers may want to go beyond the book and reader research and continue to learn by themselves. This chapter attempts to provide a foundation of the math of neural networks to allow readers without a math background do just that.
Chapter Eighteen: Bringing Everything Together
This chapter brings together what you have learned throughout the book and lays out the steps we have taken in the various projects to allow you to apply what you have learned to your own projects. Somewhat of a checklist when creating your own projects.
Chapter Nineteen: Additional Resources
A number of additional resources area provided, including: further reading, courses to tak and a list of open source data sets to use your newly developed skills.

Whats Different About PDLF?


A page highlighted showing an example diagram from the book

Designed to leave you with practical and usable skills. From the beginning you will create programs to build understanding of core concepts, ultimately moving onto larger, more complex projects.

The first exercise guides you through an implementation designed to show how basic neural networks operate and give a glimpse into what can be achieved.


Passage highlighted illustrating the approachable nature of the writing - taking complex theory and explaining in plain English

Written to be approachable to individuals from all backgrounds. Detailed explanations and illustrations provide insight into the often complex principles and methodologies of machine learning.

Where technical jargon is necessary and the readability becomes difficult, ‘Jargon Buster’ sections are used to break down the information into understandable English.

Just Enough Math

Fog circling above snowy mountains leading down to Iceberg lake surrounded by trees

You don’t need to be a math whizz to understand and use neural networks. Many of the books and videos on the subject assume advanced math skills - this book doesn’t. Math is used to illustrate concepts, but it’s explained in plain English.

One of the final chapters delves into the math of neural networks and will teach you all you need to know should you want to go deeper once you’ve finished this book.


The Ultimate Package

The Ultimate Package provides all the resources you will need to explore deep learning through practical experiments across nineteen chapters. Videos guide you through concepts and examples. Python code files allow the you to easily follow the tutorials and learn by doing. Additionally, custom software, Tweet Annotation Tool makes it easier to create your own twitter data sets for exploring Sentiment Analysis and Opinion Mining - one of the projects included.

Buy Now

The Book + Code Files

Nineteen chapters of machine learning goodness, numerous examples and tutorials, and four in-depth projects. The culmination of hundreds of hours of research, writing and design. No matter your background you will learn something new – unless you happen to have a MSc/PhD in Data Science. Includes example and project code files. You’ll use multiple open source libraries and analyse real data. You won’t regret this purchase.

Buy Now

Frequently Asked Questions

When will the book be released?

The book is scheduled for release near the end of July 2019, no hard date yet. The book has previously been delayed by my master’s study - apologies to people who have been waiting.

What format is the book provided in?

The book is provided in PDF format. Due to design considerations it is not possible to offer .mobi, .epub or other formats.

What are the differences between the packages?

There are two packages. The Book + Code Files, which is the book in PDF format along with python code files, and The Ultimate Package. The latter contains the book, python code files, the twitter annotation tool and videos inteneded to give a comprehensive introduction to deep learning. It’s really down to your budget which one is best for you.

Can I return the book if I don’t like it?

Yes. If you don’t like the book reply to your recipt email within 14 days and I will refund you. Pre-orders will be refunded 14 days from the date of launch.

I have more questions, how can I get in touch?

If you have any questions or feedback, they are greatefully recieved. Please see the contact page for specific contact information.