李沐《动手学深度学习》PyTorch 实现版开源,瞬间登上 GitHub 热榜!

李沐,亚马逊 AI 主任科学家,名声在外!半年前,由李沐、Aston Zhang 等人合力打造的《动手学深度学习》正式上线,免费供大家阅读。这是一本面向中文读者的能运行、可讨论的深度学习教科书!


李沐《动手学深度学习》PyTorch 实现版开源,瞬间登上 GitHub 热榜!


之前,红色石头就分享过这份资源,再次附上:


在线预览地址:


https://zh.d2l.ai/


GitHub 项目地址:


https://github.com/d2l-ai/d2l-zh


课程视频地址:


https://space.bilibili.com/209599371/channel/detail?cid=23541


我们知道,作为 MXNet 的作者之一,李沐的这本《动手学深度学习》也是使用 MXNet 框架写成的。但是很多入坑机器学习的萌新们使用的却是 PyTorch。如果有教材对应的 PyTorch 实现代码就更好了!


撒花!今天就给大家带来这本书的 PyTorch 实现源码。最近,来自印度理工学院的数据科学小组,把《动手学深度学习》从 MXNet “翻译”成了 PyTorch,经过 3 个月的努力,这个项目已经基本完成,并登上了 GitHub 热榜。


李沐《动手学深度学习》PyTorch 实现版开源,瞬间登上 GitHub 热榜!

首先放上这份资源的 GitHub 地址:


https://github.com/dsgiitr/d2l-pytorch


详细目录如下:


  • Ch02 Installation
    • Installation
  • Ch03 Introduction
    • Introduction
  • Ch04 The Preliminaries: A Crashcourse
    • 4.1 Data Manipulation
    • 4.2 Linear Algebra
    • 4.3 Automatic Differentiation
    • 4.4 Probability and Statistics
    • 4.5 Naive Bayes Classification
    • 4.6 Documentation
  • Ch05 Linear Neural Networks
    • 5.1 Linear Regression
    • 5.2 Linear Regression Implementation from Scratch
    • 5.3 Concise Implementation of Linear Regression
    • 5.4 Softmax Regression
    • 5.5 Image Classification Data (Fashion-MNIST)
    • 5.6 Implementation of Softmax Regression from Scratch
    • 5.7 Concise Implementation of Softmax Regression
  • Ch06 Multilayer Perceptrons
    • 6.1 Multilayer Perceptron
    • 6.2 Implementation of Multilayer Perceptron from Scratch
    • 6.3 Concise Implementation of Multilayer Perceptron
    • 6.4 Model Selection Underfitting and Overfitting
    • 6.5 Weight Decay
    • 6.6 Dropout
    • 6.7 Forward Propagation Backward Propagation and Computational Graphs
    • 6.8 Numerical Stability and Initialization
    • 6.9 Considering the Environment
    • 6.10 Predicting House Prices on Kaggle
  • Ch07 Deep Learning Computation
    • 7.1 Layers and Blocks
    • 7.2 Parameter Management
    • 7.3 Deferred Initialization
    • 7.4 Custom Layers
    • 7.5 File I/O
    • 7.6 GPUs
  • Ch08 Convolutional Neural Networks
    • 8.1 From Dense Layers to Convolutions
    • 8.2 Convolutions for Images
    • 8.3 Padding and Stride
    • 8.4 Multiple Input and Output Channels
    • 8.5 Pooling
    • 8.6 Convolutional Neural Networks (LeNet)
  • Ch09 Modern Convolutional Networks
    • 9.1 Deep Convolutional Neural Networks (AlexNet)
    • 9.2 Networks Using Blocks (VGG)
    • 9.3 Network in Network (NiN)
    • 9.4 Networks with Parallel Concatenations (GoogLeNet)
    • 9.5 Batch Normalization
    • 9.6 Residual Networks (ResNet)
    • 9.7 Densely Connected Networks (DenseNet)
  • Ch10 Recurrent Neural Networks
    • 10.1 Sequence Models
    • 10.2 Language Models
    • 10.3 Recurrent Neural Networks
    • 10.4 Text Preprocessing
    • 10.5 Implementation of Recurrent Neural Networks from Scratch
    • 10.6 Concise Implementation of Recurrent Neural Networks
    • 10.7 Backpropagation Through Time
    • 10.8 Gated Recurrent Units (GRU)
    • 10.9 Long Short Term Memory (LSTM)
    • 10.10 Deep Recurrent Neural Networks
    • 10.11 Bidirectional Recurrent Neural Networks
    • 10.12 Machine Translation and DataSets
    • 10.13 Encoder-Decoder Architecture
    • 10.14 Sequence to Sequence
    • 10.15 Beam Search
  • Ch11 Attention Mechanism
    • 11.1 Attention Mechanism
    • 11.2 Sequence to Sequence with Attention Mechanism
    • 11.3 Transformer
  • Ch12 Optimization Algorithms
    • 12.1 Optimization and Deep Learning
    • 12.2 Convexity
    • 12.3 Gradient Descent
    • 12.4 Stochastic Gradient Descent
    • 12.5 Mini-batch Stochastic Gradient Descent
    • 12.6 Momentum
    • 12.7 Adagrad
    • 12.8 RMSProp
    • 12.9 Adadelta
    • 12.10 Adam


其中,每一小节都是可以运行的 Jupyter 记事本,你可以*修改代码和超参数来获取及时反馈,从而积累深度学习的实战经验。


目前,PyTorch 代码还有 6 个小节没有完成,但整体的完成度已经很高了!开发团队希望更多的爱好者加入进来,贡献一份力量!


最后,再次附上 GitHub 地址:


https://github.com/dsgiitr/d2l-pytorch

上一篇:ecs学生服务器搭建discuz初体验


下一篇:转载 - C++ bitset类使用与简介