This file has been truncated. show original
# How to use this book
This book will provide a comprehensive introduction to all aspects of deep learning from model construction to model training, as well as their applications in computer vision and natural language processing. We will not only explain the principles of the algorithm, but also demonstrate their implementation and operation based on Apache MXNet. Each section of the book is a Jupyter notebook, which combines texts, formulas, images, codes, and running results. Not only can you read them directly, but you can run them to get an interactive learning experience.
## Target reader
This book is for college students, engineers, and researchers who wish to learn deep learning, especially for those who are interested in applying deep learning into practice. This book does not require that you have any background in deep learning or machine learning. We will explain every concept from scratch. Although illustrations of deep learning techniques and applications involve mathematics and programming, you only need to know their basics, such as basic linear algebra, calculus, and probability, and basic Python programming. In appendix we provide most of the mathematics covered in this book for your reference. If you have not used Python before, you may refer to the tutorial http://learnpython.org/ . Of course, if you are only interested in the mathematical part, you can ignore the programming part, and vice versa.
## Content and structure
The book can be roughly divided into three sections:
* The first section (Chapters 1–3) covers prerequisite work and basic knowledge concerned. Chapter 1 introduces the background of deep learning and how to use this book. Chapter 2 provides the prerequisites required for hands-on deep learning, such as how to acquire and run the codes covered in the book. Chapter 3 covers the most basic concepts and techniques of deep learning, such as multi-layer perceptrons and model regularization. If time does not permit or you only want to learn about the most basic concepts and techniques of deep learning, then it suffices to read the first section only.
* The second section (Chapters 4-6) focuses on modern deep learning techniques. Chapter 4 describes the various key components of deep learning calculations and lays the groundwork for the later implementation of more complex models. Chapter 5 explains the convolutional neural networks that have made deep learning a success in computer vision in recent years. Chapter 6 describes the recurrent neural networks that are commonly used to process sequence data in recent years. Reading through the second section will help you grasp modern deep learning techniques.
* The third section (Chapters 7-10) discusses computing performance and applications. Chapter 7 evaluates various optimization algorithms used to train deep learning models. Chapter 8 examines several important factors that affect the performance of deep learning computation. Chapters 9 and 10 respectively illustrate the major applications of deep learning in computer vision and natural language processing. This section is for you to optionally read based on your interests.
Figure 1.2 outlines the structure of the book.