Deep Learning with PyTorch Chinese

Deep-Learning-with-PyTorch-Chinese

This project (web version portal) translates the official PyTorch book "Deep learning with PyTorch" (basic excerpt version) into Chinese and gives the relevant code that can be runnable throughout the book.

This project translates the PyTorch official book "Deep learning with PyTorch" (essential excerpt version) into Chinese.

1. Book introduction

Since its inception in 2016, PyTorch has become one of the hottest deep learning frameworks today. Recently, the official authoritative PyTorch tutorial book "Deep learning with PyTorch" has finally been released. As soon as the news came out, it was highly recommended by the giant Yann LeCun. It is an excellent textbook for getting started with PyTorch and deep learning.

It should be noted that the PDF provided by PyTorch official website is a basic excerpt (Essential Excelpts), with a total of 141 pages, and the content includes the following five parts:

1. Introduction to Deep Learning and PyTorch
2. Start with a tensor
3. Use tensors to represent real data
4. Learning mechanism
5. Fitting data using neural network

Therefore, it can be used as a tutorial for getting started with PyTorch. The full version of this book is also available for free preview, portal.

2. Project Introduction

This project translates the original book into Chinese and gives relevant codes that can be runnable.

This repository mainly contains two folders: code and docs (plus some data is stored in data). The code folder is the relevant jupyter notebook code for each chapter; the docs folder is the Chinese translation of the relevant content in the markdown format "Deep learning with PyTorch" (basic excerpt version), and then use docsify to deploy the web document to GitHub Pages. Contributions to this project or issue is welcome.

3. How to use

This project is aimed at children's shoes that are interested in PyTorch, especially those who want to get started quickly. This project does not require you to have any background knowledge of deep learning or machine learning. You only need to understand basic mathematics and programming, such as basic linear algebra, differential and probability, and basic Python programming.

The documentation of this repository contains some latex formulas, but the markdown native of github does not support the display of formulas, and the docs folder has been deployed on GitHub Pages using docsify, so you can easily access the web version of this project. If you want to run the relevant code, you need to clone the project and then run the relevant code in the code folder.

Since the basic excerpts translated in this project, it is only 141 pages, it is suitable for a quick start with PyTorch. If you want to have more in-depth learning of PyTorch and deep learning (such as computer vision, natural language processing, etc.), you may need more information. If you are interested, you can refer to another project of my Dive-into-DL-PyTorch.

4. Contents

• Introduction
• 1. Introduction to Deep Learning and PyTorch
  • 1.1 What is PyTorch
  • 1.2 What does this book tell about
  • 1.3 Why use PyTorch
  • 1.4 Components included in PyTorch
• 2. Start with a tensor
  • 2.1 Tensor Basics
  • 2.2 Tensors and Storage
  • 2.3 Size, storage offset and step size
  • 2.4 Data Type
  • 2.5 Index Tensor
  • 2.6 Intercommunication with NumPy
  • 2.7 Serialize Tensors
  • 2.8 Transfer tensors to GPU to run
  • 2.9 Tensor API
• 3. Use tensors to represent real data
  • 3.1 Tabular data
  • 3.2 Time series
  • 3.3 Text data
  • 3.4 Image Data
  • 3.5 Volume data
• 4. Learning mechanism
  • 4.1 Learning is parameter estimation
  • 4.2 PyTorch automatic guidance
• 5. Use neural network to fit data
  • 5.1 Neurons
  • 5.2 PyTorch's nn module
  • 5.3 nn subclass

5. Statement

• The translator translates purely for learning purposes and personal interests, and does not pursue any economic benefits;
• This project is limited to the use of learning and research purposes only. The translator reserves the right to name the project. Any reprint must indicate the source, but shall not be used for any commercial purpose;
• The use of this project infringement of the original work or any act that violates the Intellectual Property Protection Law has nothing to do with the translator;
• Those who are able to read English books should read the original version or purchase the full version of the book.

LICENSE

CC BY-NC (Attribution-Non-Commercial Use) 4.0