Deep learning can train machines to perform incredible tasks , For example, face recognition , Cancer detection , Even stock market forecasts . That's why we need to learn deep learning .

Here are some books you can learn from scratch . Each book in this review has its own advantages , And every book is excellent . I've ranked them in the order I think are the best and the order in which I suggest learning .

1. use Scikit-Learn and TensorFlow Machine learning

The author tries to explain complex topics in a way that almost anyone can understand , and , You can see how to deal with real data , How to visualize data for insight , And it's important to prepare data for machine learning algorithms .

It's at the back of the book , You will see the famous MNIST classifier , How to train the model and some basic machine learning classifiers , as SVM, Decision tree , Random forest, etc .

All of this is to get you ready for the second part of the book , The content involves Tensorflow( Including installation ) And basic neural network and deep neural network .

I think the structure of the book is very good , The topics are introduced in the correct order , And the ideas and examples in the book are well explained .

2. Deep learning (Deep Learning)

This book is regarded by many as the Bible of deep learning , Because it brings together years and years of learning and focused research in a book .

This book is not meant to concentrate on study , It can be better used for bedtime reading , Because it's full of functional equations , And written in typical textbooks , So it won't be written in the most interesting style .

It introduces basic mathematics from the beginning , Such as linear algebra , probability theory , Then we turn to machine learning basics , Finally, it introduces deep network and deep learning .

If you are an aspiring student who is eager to master the subject and enter into deep learning and research , Then this book will certainly help you . This is probably the most comprehensive book on the subject so far .

3.Deep Learning for the Layman( Deep learning for laymen )

As the title says , It is written for the general reader .

For the deep learning of the layman, we first introduce the deep learning , say concretely , What is it and why it is needed .

The next part of the book explains supervised learning , The difference between unsupervised learning and reinforcement learning , Classification and clustering are also introduced . Artificial neural networks will be discussed later in this book , It includes how they are built and the parts that make up each layer of the network . Finally, the deep learning is discussed , It includes convolutional neural networks that form part of many computer vision algorithms today .

If you want a simple English guide , At the same time, we can see the words that are rarely hyped , Then this book may be for you .

4. Build your own neural network (Make Your Own Neural Network)

It's not strictly “ Deep learning ”, But this book will give you a deeper understanding of neural networks and how they work , Help you understand deep neural networks .

In this book , You can do math with neural networks , Fully understand the working mode of neural network . You can not only know how they work , You can also use Python Two examples of neural network implementation in , This will help consolidate your understanding of the subject .

This book begins with a high-level overview of machine learning , Then the details of neural network are studied . The mathematics involved did not exceed the university level , But it includes an introduction to calculus , This is explained in the way that as many people as possible visit .

5. Deep learning beginners (Deep Learning for Beginners)

Deep learning for beginners , This book doesn't pay much attention to the mathematics of deep learning , Instead, use charts to help you understand the basic concepts and algorithms of deep learning .

This book takes a different approach from many other books , By providing a simple example of how the deep learning algorithm works , These examples are then built step by step and more complex algorithmic parts are introduced .

In terms of book structure , You will first learn the basics of artificial neural networks , And understand the difference between machine learning and deep learning . after , You're going into convolutional neural networks (CNN) And other deep learning algorithms before learning about multilayer perceptron (MLP) All information for .

This is a good book for beginners , These concepts can be well explained , But if you're looking for something more practical , Then you should look for other books in this review .

6. Neural network and deep learning (Neural Networks and Deep Learning:Deep Learning explained to your

A popular book on deep learning , In short , Your grandmother can understand deep learning with the help of this book !

Neural network and deep learning : Let you gradually understand the basic knowledge of neural network and deep learning , For those who want to understand the subject, but don't necessarily want to get to know all the math backgrounds , This book is a great book .

therefore , After a brief introduction to machine learning , You will learn supervised learning and unsupervised learning , And then study neurons like , Activation function and different types of network architecture .

last , You will learn how deep learning actually works , Main types of deep neural networks ( Including convolution neural network ), How to provide memory for neural network , The various frameworks and libraries available are also discussed .

7. Foundation of deep learning : Design next generation machine intelligence algorithm (Fundamentals of Deep Learning: Designing Next-Generation
Machine Intelligence Algorithms)

Nikhil Buduma and Nicholas Locascio This book was written and designed to help you start deep learning , But the goal is for those familiar with Python And people with a background in calculus .

I think one of the highlights of this book is that it's heavily used Tensorflow, It is Google Deep learning framework of , It is used to construct neural network . in fact , There is a whole chapter devoted to it , This, in my opinion, is a huge advantage .

For the rest of the book , It involves some fairly advanced features , If the gradient drops , Convolution filter , Deep intensive learning and so on .

8. study TensorFlow: A guide to building a deep learning system (Learning TensorFlow: A Guide to Building Deep
Learning Systems)

Next is a full attention book Tensorflow The book of , This book is for Tensorflow A practical method is provided , Suitable for a wide range of technical personnel , From data scientist to engineer , student .

Through the Tensorflow Some basic examples are provided in , This book is very introductory at the beginning , But then it turned to deeper themes , Such as convolution neural network and other neural network architecture , How to use text and sequences ,TensorBoard visualization ,TensorFlow Abstract library and multi thread input pipeline .

study TensorFlow The ultimate goal is to teach you how to save and export models and how to use them Tensorflow service API, stay Tensorflow Build and deploy a deep learning system suitable for production .

9. use Python Deep learning (Deep Learning with Python)

Deep learning with python As a title, it is suggested to introduce the use of deep learning Python Programming language and open source Keras library , It allows simple and rapid prototyping .

stay Python In depth learning ,   You will learn deep learning from the beginning , You will learn all about image classification models , How to use deep learning to get text and sequence , You can even learn how to use neural networks to generate text and images .

This book is for those with Python Written by skilled personnel , But you don't have to learn on the machine ,Tensorflow or Keras Have any experience with . You don't need advanced math background either , Only basic high school level mathematics should allow you to follow and understand the core ideas .

10. Deep learning : Practitioners' approach (Deep Learning :A Practitioner’s Approach)

This book focuses on the introduction Deep Learning For Java(DL4J), It is used to train and implement deep neural networks Java frame / library .

Now most AI research is using Python Ongoing , Because rapid prototyping is usually faster , But as more organizations ( Many of them use Java) embrace AI, We may see more AI Turning algorithm Java, as DL4J.

This book is a beginner's book about deep learning , But if you already have it Java Or deep learning experience , So you can go straight to the example .

By reading this book , You will understand the concept of machine learning in general , Pay special attention to deep learning . You'll learn how deep neural networks evolved from basic neural networks , You'll also learn about some deep network architectures , Such as convolution neural network and cyclic neural network .

If you know Hadoop and Spark, Then you will be able to understand how to use it DL4J The technology itself .

11. use TensorFlow Professional deep learning (Pro Deep Learning with TensorFlow)

This book will teach you in a hands-on way Tensorflow, So that you can learn from scratch, deep learning , Quick Mastery Tensorflow API And learn how to optimize various deep learning network architecture .

Professional deep learning will help you develop the mathematical knowledge and experience needed to adjust the existing neural network architecture , Even create a new architecture that may challenge the latest technology level .

All the codes in this book use iPython In the form of notebook , Because I used it in the past Tensorflow, I found that using iPython Laptops are very useful .

This book is for data scientists and machine learning professionals , Software developers , Graduate students and open source enthusiasts , And will provide you with mathematical basis and machine learning principles , Enables you to conduct research and deploy deep learning solutions to production environments .

12. For deep learning TensorFlow(TensorFlow for Deep Learning)

This book will introduce you to the concept of deep learning through examples from scratch , Specially designed for experienced developers who build software systems, but have no experience in deep learning architecture .

This book will show you how to design executable object detection , Translation of human language , Systems that analyze videos and even predict potential drug properties !

You will get information about Tensorflow API In depth knowledge of , How to train neural network on large data set and how to train convolution network , Circular network ,LSTM And reinforcement learning TensorFlow.

This book does need some background knowledge of basic linear algebra and calculus , But it's a practical book , It's designed to teach you how to create a system that you can learn .

artificial intelligence , big data , The future development of cloud computing and Internet of things is worthy of attention , They are all frontier industries , Interested friends , Look at the age of wisdom , Here for you to recommend a few good quality articles :

* Optimizing the calculation of deep learning , What are the main challenges ? <>
* Deep learning , Machine learning and NLP, Why learn these new technologies <>
* The relationship between deep learning and data center <>