DeepLearning MuLi Notes Download - DeepLearning MuLi Notes Source code download

DeepLearning MuLi Notes

Python

1.0.0

Download

"Hand-On Learning Deep Learning"

Project motivation/ Course introduction/ Course resources/ Textbook catalog/ Notes/ Folder description/ Organizer/ Contributor

Project motivation

"Hand-On Learning Deep Learning" is a series of deep learning videos taught by Mr. Li Mu (a senior chief scientist at AWS, Ph.D., Department of Computer Science, Carnegie Mellon University, USA). This project collects detailed markdown notes and related jupyter codes during our winter vacation learning "Hand-On Learning Deep Learning" process. Giving someone a rose will leave a fragrance on your hands. We open source all markdown notes. We hope that while learning by ourselves, it will also be helpful for everyone to learn and master Teacher Li Mu's "Hand-On Learning Deep Learning".

Features of this project:
Markdown notes correspond to the original course video one by one, which can help everyone understand while listening to the class.
The jupyter codes have detailed Chinese annotations to help everyone get started with practice faster.

There are 73 course videos in total , and the average duration of a single video is no more than 30 minutes . It is expected that the study can be completed within 40 days of winter vacation .

The badge used in this project comes from the Internet. If it infringes on your image copyright, please contact us to delete it. Thank you.

Course Introduction

Usually when we mention deep learning, we often forget that deep learning is just a small part of machine learning, and think it is a separate module independent of machine learning. This is because machine learning, as a discipline with a longer history, had a narrow range of applications in the real world before deep learning was released. In the fields of speech recognition, computer vision, natural language processing, etc., machine learning is often just a small part of the solutions to solve these domain problems because of the large amount of domain knowledge required and the reality is extremely complex. But in the past few years, the advent and application of deep learning have brought surprises to the world, promoting the rapid development of computer vision, natural language processing, automatic speech recognition, reinforcement learning and statistical modeling, and gradually leading the trend, setting off a wave of artificial intelligence revolution in the world.

In the course "Hand-On Deep Learning" , there are not only a small amount of basic knowledge of machine learning, such as linear neural networks, multi-layer perception machines , etc.; there are also various deep learning models that are currently used in cutting-edge applications: including leNet, ResNet, LSTM, BERT... At the same time, the explanation of each chapter is also equipped with code, textbooks, etc. implemented by pytorch, which can help students master the basic models and cutting-edge knowledge of deep learning in a short period of time and improve their practical ability.

Course Resources

Bilibili teaching website: hands-on learning deep learning (completed)
Textbook URL: "Hand-On Deep Learning" (Chinese version), "Dive into Deep Learning" 0.17.1 documentation
Course homepage: https://courses.d2l.ai/zh-v2/
Forum URL: discuss.d2l.ai, discuss.pytorch

In addition, this course has corresponding code implementations. Each chapter has a corresponding jupyter notepad, providing the complete python code of the model , and all resources can be obtained for free online.

Textbook catalog

"Hand-On Deep Learning" (Chinese version) and English version Dive into Deep Learning 0.17.1 documentation textbook catalog and chapter links are as follows:

chapter	Chinese version	English version
1	Preface	Introduction
2	Preparation knowledge	Preliminaries
3	Linear neural network	Linear Neural Networks
4	Multi-layer sensing machine	Multilayer Perceptrons
5	Deep Learning Computing	Deep Learning Computation
6	Convolutional neural network	Convolutional Neural Networks
7	Modern convolutional neural networks	Modern Convolutional Neural Networks
8	Recurrent neural network	Recurrent Neural Networks
9	Modern recurrent neural network	Modern Recurrent Neural Networks
10	Attention mechanism	Attention Mechanisms
11	Optimization algorithm	Optimization Algorithms
12	Computational performance	Computational Performance
13	Computer Vision	Computer Vision
14	Natural Language Processing: Pre-training	Natural Language Processing: Pretraining
15	Natural Language Processing: Application	Natural Language Processing Applications
16	Recommended system	Recommender Systems
17	Generate adversarial networks	Generative Adversarial Networks
18	Appendix: The basics of mathematics in deep learning	Appendix: Mathematics for Deep Learning
19	Appendix: Deep Learning Tools	Appendix: Tools for Deep Learning

notes

notes	Handout	Code
00-Preview	Handout
01-Course Schedule	Handout
02-Introduction to Deep Learning	Handout
03-Installation	Handout
04-Data operation and data preprocessing	Handout	Jupyter Code
05-Linear Algebra	Handout
06-Matrix calculation	Handout
07-Chengdu Rules and Automatic Derivation	Handout	Jupyter Code
08-Linear Regression + Basic Optimization Algorithm	Lecture Notes 1 2	Jupyter Code
09-Softmax Regression	Lecture Notes 1 2	Jupyter Code
10-Multi-layer sensing machine	Lecture Notes 1 2	Jupyter Code
11-Model selection + overfitting and underfitting	Lecture Notes 1 2
12-Weight Recession	Handout	Jupyter Code
13-Discarding method	Handout	Jupyter Code
14-Numerical Stability	Lecture Notes 1 2
15-Practical Kaggle Prediction of House Prices	Handout	Jupyter Code
16-Pytorch neural network basics	Handout	Jupyter Code
17-Use and Buy GPUs	Handout	Jupyter Code
18-Summary of the Prediction House Price Competition	Handout
19-Convolutional Layer	Lecture Notes 1 2	Jupyter Code
20-Fill and stride	Handout	Jupyter Code
21-Multi-input and output channels	Handout	Jupyter Code
22-Pooling Layer	Handout	Jupyter Code
23-LeNet	Handout	Jupyter Code
24-AlexNet	Handout	Jupyter Code
25-Network VGG using blocks	Handout	Jupyter Code
26-NiN	Handout	Jupyter Code
27-GoogLeNet	Handout	Jupyter Code
28-Batch Normalization	Handout	Jupyter Code
29-Residual Network ResNet	Handout
30-Part 2 Finish the Competition	Handout
31-CPU and GPU	Handout
32-Deep Learning Hardware	Handout
33-Single-machine multiple cards in parallel	Handout
34-Multi-GPU training implementation	Handout	Jupyter Code
35-Distributed Training	Handout
36-Data augmentation	Handout	Jupyter Code
37-fine adjustment	Handout	Jupyter Code
38-The results of the second competition leaf classification	Handout
39-Practical Kaggle Competition-1	Handout
40-Practical Kaggle Competition-2	Handout	Jupyter Code
41-Object Detection and Data Sets	Handout
[42-Anchor Frame]	Handout
43-Technical summary of leaves classification competition	Handout
44-Object Detection Algorithm	Lecture Notes 123
45-SSD implementation	Handout	Jupyter Code
46-Semantic segmentation and datasets	Handout	Jupyter Code
47-Transfer Convolution	Lecture Notes 1 2
48-FCN	Handout	Jupyter Code
49-Style Migration	Handout	Jupyter Code
50-Course Competition	Handout
51-Sequence Model	Handout
52-Text Preprocessing	Handout	Jupyter Code
53-Language Model	Handout
54-Recurrent Neural Network	Handout
55-RNN implementation	Handout	Jupyter Code
56-GRU	Handout	Jupyter Code
57-LSTM	Handout
58-Deep Recurrent Neural Network	Handout	Jupyter Code
[59-Bidirectional recurrent neural network]	Handout
60-Machine Translation Dataset	Handout	Jupyter Code
61-Encoder-Decoder Architecture	Handout
62-Sequence-to-Sequence Learning	Handout	Jupyter Code
63-Bunch Search	Handout
64- Attention Mechanism	Handout	Jupyter Code
65- Attention Score	Handout
[66-seq using attention]	Handout
67-Self-attention	Handout	Jupyter Code
68-Transformer	Handout	Jupyter Code
69-BERT pre-training	Handout
70-BERT fine adjustment	Handout
[71-Summary of the Target Detection Competition]	Handout
72-Optimization Algorithm	Handout
73-Course Summary and Advanced Learning	Handout