一网打尽！深度学习 100+ 经典模型 TensorFlow 与 Pytorch 代码实现大集合

转自 | 专知

深度学习在过去十年获得了极大进展，出现很多新的模型，并且伴随TensorFlow和Pytorch框架的出现，有很多实现，但对于初学者和很多从业人员，如何选择合适的实现，是个选择。rasbt在Github上整理了关于深度学习模型TensorFlow和Pytorch代码实现集合，含有100个，各种各样的深度学习架构，模型，和技巧的集合Jupyter Notebooks，从基础的逻辑回归到神经网络到CNN到GNN等，可谓一网打尽，值得收藏！

项目地址：

https://github.com/rasbt/deeplearning-models

传统机器学习

感知器 Perceptron
[TensorFlow 1: GitHub | Nbviewer]
https://github.com/rasbt/deeplearning-models/blob/master/tensorflow1_ipynb/basic-ml/perceptron.ipynb
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/basic-ml/perceptron.ipynb

逻辑回归 Logistic Regression
[TensorFlow 1: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/tensorflow1_ipynb/basic-ml/logistic-regression.ipynb
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/basic-ml/logistic-regression.ipynb

Softmax Regression (Multinomial Logistic Regression)
[TensorFlow 1: GitHub | Nbviewer]
https://github.com/rasbt/deeplearning-models/blob/master/tensorflow1_ipynb/basic-ml/softmax-regression.ipynb
[PyTorch: GitHub | Nbviewer]
https://github.com/rasbt/deeplearning-models/blob/master/pytorch_ipynb/basic-ml/softmax-regression.ipynb

Softmax Regression with MLxtend's plot_decision_regions on Iris
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/basic-ml/softmax-regression-mlxtend-1.ipynb

多层感知器

多层感知器 Multilayer Perceptron
[TensorFlow 1: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/tensorflow1_ipynb/mlp/mlp-basic.ipynb
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/mlp/mlp-basic.ipynb

带Dropout的多层感知器 Multilayer Perceptron with Dropout
[TensorFlow 1: GitHub | Nbviewer]
[PyTorch: GitHub | Nbviewer]
具有批处理规范化的多层感知器 Multilayer Perceptron with Batch Normalization
[TensorFlow 1: GitHub | Nbviewer]
[PyTorch: GitHub | Nbviewer]
Multilayer Perceptron with Backpropagation from Scratch
[TensorFlow 1: GitHub | Nbviewer]
[PyTorch: GitHub | Nbviewer]

卷积神经网络

基础

卷积神经网络 Convolutional Neural Network
[TensorFlow 1: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/tensorflow1_ipynb/cnn/cnn-basic.ipynb
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/cnn/cnn-basic.ipynb
Convolutional Neural Network with He Initialization
[PyTorch: GitHub | Nbviewer]

Concepts

Replacing Fully-Connnected by Equivalent Convolutional Layers
[PyTorch: GitHub | Nbviewer]

Fully Convolutional

Fully Convolutional Neural Network
[PyTorch: GitHub | Nbviewer]

LeNet

LeNet-5 on MNIST
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/cnn/cnn-lenet5-mnist.ipynb
LeNet-5 on CIFAR-10
[PyTorch: GitHub | Nbviewer]
LeNet-5 on QuickDraw
[PyTorch: GitHub | Nbviewer]

AlexNet

AlexNet on CIFAR-10
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/cnn/cnn-alexnet-cifar10.ipynb

VGG

Convolutional Neural Network VGG-16
[TensorFlow 1: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/tensorflow1_ipynb/cnn/cnn-vgg16.ipynb
[PyTorch: GitHub | Nbviewer]
VGG-16 Gender Classifier Trained on CelebA
[PyTorch: GitHub | Nbviewer]
Convolutional Neural Network VGG-19
[PyTorch: GitHub | Nbviewer]

DenseNet

DenseNet-121 Digit Classifier Trained on MNIST
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/cnn/cnn-densenet121-mnist.ipynb
DenseNet-121 Image Classifier Trained on CIFAR-10
[PyTorch: GitHub | Nbviewer]

ResNet

ResNet and Residual Blocks
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/cnn/resnet-ex-1.ipynb
ResNet-18 Digit Classifier Trained on MNIST
[PyTorch: GitHub | Nbviewer]
ResNet-18 Gender Classifier Trained on CelebA
[PyTorch: GitHub | Nbviewer]
ResNet-34 Digit Classifier Trained on MNIST
[PyTorch: GitHub | Nbviewer]
ResNet-34 Object Classifier Trained on QuickDraw
[PyTorch: GitHub | Nbviewer]
ResNet-34 Gender Classifier Trained on CelebA
[PyTorch: GitHub | Nbviewer]
ResNet-50 Digit Classifier Trained on MNIST
[PyTorch: GitHub | Nbviewer]
ResNet-50 Gender Classifier Trained on CelebA
[PyTorch: GitHub | Nbviewer]
ResNet-101 Gender Classifier Trained on CelebA
[PyTorch: GitHub | Nbviewer]
ResNet-101 Trained on CIFAR-10
[PyTorch: GitHub | Nbviewer]
ResNet-152 Gender Classifier Trained on CelebA
[PyTorch: GitHub | Nbviewer]

Network in Network

Network in Network CIFAR-10 Classifier
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/cnn/nin-cifar10.ipynb

归一化层 Normalization Layers

BatchNorm before and after Activation for Network-in-Network CIFAR-10 Classifier
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/cnn/nin-cifar10_batchnorm.ipynb
Filter Response Normalization for Network-in-Network CIFAR-10 Classifier
[PyTorch: GitHub | Nbviewer]

度量学习 Metric Learning

Siamese Network with Multilayer Perceptrons
[TensorFlow 1: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/tensorflow1_ipynb/metric/siamese-1.ipynb

自编码器 Autoencoders

全连接自编码器 Fully-connected Autoencoders

Autoencoder (MNIST)
[TensorFlow 1: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/tensorflow1_ipynb/autoencoder/ae-basic.ipynb
[PyTorch: GitHub | Nbviewer]
Autoencoder (MNIST) + Scikit-Learn Random Forest Classifier
[TensorFlow 1: GitHub | Nbviewer]
[PyTorch: GitHub | Nbviewer]

Convolutional Autoencoders

Convolutional Autoencoder with Deconvolutions / Transposed Convolutions
[TensorFlow 1: GitHub | Nbviewer]
[PyTorch: GitHub | Nbviewer]
Convolutional Autoencoder with Deconvolutions and Continuous Jaccard Distance
[PyTorch: GitHub | Nbviewer]
Convolutional Autoencoder with Deconvolutions (without pooling operations)
[PyTorch: GitHub | Nbviewer]
Convolutional Autoencoder with Nearest-neighbor Interpolation
[TensorFlow 1: GitHub | Nbviewer]
[PyTorch: GitHub | Nbviewer]
Convolutional Autoencoder with Nearest-neighbor Interpolation -- Trained on CelebA
[PyTorch: GitHub | Nbviewer]
Convolutional Autoencoder with Nearest-neighbor Interpolation -- Trained on Quickdraw
[PyTorch: GitHub | Nbviewer]

Variational Autoencoders

Variational Autoencoder
[PyTorch: GitHub | Nbviewer]
Convolutional Variational Autoencoder
[PyTorch: GitHub | Nbviewer]

Conditional Variational Autoencoders

Conditional Variational Autoencoder (with labels in reconstruction loss)
[PyTorch: GitHub | Nbviewer]
Conditional Variational Autoencoder (without labels in reconstruction loss)
[PyTorch: GitHub | Nbviewer]
Convolutional Conditional Variational Autoencoder (with labels in reconstruction loss)
[PyTorch: GitHub | Nbviewer]
Convolutional Conditional Variational Autoencoder (without labels in reconstruction loss)
[PyTorch: GitHub | Nbviewer]

生成式对抗网络 (GANs)

Fully Connected GAN on MNIST
[TensorFlow 1: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/tensorflow1_ipynb/gan/gan.ipynb
[PyTorch: GitHub | Nbviewer]
Fully Connected Wasserstein GAN on MNIST
[PyTorch: GitHub | Nbviewer]
Convolutional GAN on MNIST
[TensorFlow 1: GitHub | Nbviewer]
[PyTorch: GitHub | Nbviewer]
Convolutional GAN on MNIST with Label Smoothing
[TensorFlow 1: GitHub | Nbviewer]
[PyTorch: GitHub | Nbviewer]
Convolutional Wasserstein GAN on MNIST
[PyTorch: GitHub | Nbviewer]

图神经网络 (GNNs)

Most Basic Graph Neural Network with Gaussian Filter on MNIST
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/gnn/gnn-basic-1.ipynb
Basic Graph Neural Network with Edge Prediction on MNIST
[PyTorch: GitHub | Nbviewer]
Basic Graph Neural Network with Spectral Graph Convolution on MNIST
[PyTorch: GitHub | Nbviewer]

循环神经网络 (RNNs)

Many-to-one: Sentiment Analysis / Classification

A simple single-layer RNN (IMDB)
[PyTorch: GitHub | Nbviewer]
https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/rnn/rnn_simple_imdb.ipynb
A simple single-layer RNN with packed sequences to ignore padding characters (IMDB)
[PyTorch: GitHub | Nbviewer]
RNN with LSTM cells (IMDB)
[PyTorch: GitHub | Nbviewer]
RNN with LSTM cells (IMDB) and pre-trained GloVe word vectors
[PyTorch: GitHub | Nbviewer]
RNN with LSTM cells and Own Dataset in CSV Format (IMDB)
[PyTorch: GitHub | Nbviewer]
RNN with GRU cells (IMDB)
[PyTorch: GitHub | Nbviewer]
Multilayer bi-directional RNN (IMDB)
[PyTorch: GitHub | Nbviewer]
Bidirectional Multi-layer RNN with LSTM with Own Dataset in CSV Format (AG News)
[PyTorch: GitHub | Nbviewer]
Bidirectional Multi-layer RNN with LSTM with Own Dataset in CSV Format (Yelp Review Polarity)
[PyTorch: GitHub | Nbviewer]
Bidirectional Multi-layer RNN with LSTM with Own Dataset in CSV Format (Amazon Review Polarity)
[PyTorch: GitHub | Nbviewer]

Many-to-Many / Sequence-to-Sequence

A simple character RNN to generate new text (Charles Dickens)
[PyTorch: GitHub | Nbviewer]

Ordinal Regression

Ordinal Regression CNN -- CORAL w. ResNet34 on AFAD-Lite
[PyTorch: GitHub | Nbviewer]
Ordinal Regression CNN -- Niu et al. 2016 w. ResNet34 on AFAD-Lite
[PyTorch: GitHub | Nbviewer]
Ordinal Regression CNN -- Beckham and Pal 2016 w. ResNet34 on AFAD-Lite
[PyTorch: GitHub | Nbviewer]

Tips and Tricks

Cyclical Learning Rate
[PyTorch: GitHub | Nbviewer]
Annealing with Increasing the Batch Size (w. CIFAR-10 & AlexNet)
[PyTorch: GitHub | Nbviewer]
Gradient Clipping (w. MLP on MNIST)
[PyTorch: GitHub | Nbviewer]

迁移学习 Transfer Learning

Transfer Learning Example (VGG16 pre-trained on ImageNet for Cifar-10)

[PyTorch: GitHub | Nbviewer

https://nbviewer.jupyter.org/github/rasbt/deeplearning-models/blob/master/pytorch_ipynb/transfer/transferlearning-vgg16-cifar10-1.ipynb

PyTorch Workflows and Mechanics

Custom Datasets

Custom Data Loader Example for PNG Files
[PyTorch: GitHub | Nbviewer]
Using PyTorch Dataset Loading Utilities for Custom Datasets -- CSV files converted to HDF5
[PyTorch: GitHub | Nbviewer]
Using PyTorch Dataset Loading Utilities for Custom Datasets -- Face Images from CelebA
[PyTorch: GitHub | Nbviewer]
Using PyTorch Dataset Loading Utilities for Custom Datasets -- Drawings from Quickdraw
[PyTorch: GitHub | Nbviewer]
Using PyTorch Dataset Loading Utilities for Custom Datasets -- Drawings from the Street View House Number (SVHN) Dataset
[PyTorch: GitHub | Nbviewer]
Using PyTorch Dataset Loading Utilities for Custom Datasets -- Asian Face Dataset (AFAD)
[PyTorch: GitHub | Nbviewer]
Using PyTorch Dataset Loading Utilities for Custom Datasets -- Dating Historical Color Images
[PyTorch: GitHub | Nbviewer]

Training and Preprocessing

Generating Validation Set Splits
[PyTorch]: GitHub | Nbviewer]
Dataloading with Pinned Memory
[PyTorch: GitHub | Nbviewer]
Standardizing Images
[PyTorch: GitHub | Nbviewer]
Image Transformation Examples
[PyTorch: GitHub | Nbviewer]
Char-RNN with Own Text File
[PyTorch: GitHub | Nbviewer]
Sentiment Classification RNN with Own CSV File
[PyTorch: GitHub | Nbviewer]

Parallel Computing

Using Multiple GPUs with DataParallel -- VGG-16 Gender Classifier on CelebA
[PyTorch: GitHub | Nbviewer]

Other

Sequential API and hooks
[PyTorch: GitHub | Nbviewer]
Weight Sharing Within a Layer
[PyTorch: GitHub | Nbviewer]
Plotting Live Training Performance in Jupyter Notebooks with just Matplotlib
[PyTorch: GitHub | Nbviewer]

Autograd

Getting Gradients of an Intermediate Variable in PyTorch
[PyTorch: GitHub | Nbviewer]

TensorFlow Workflows and Mechanics

Custom Datasets

Chunking an Image Dataset for Minibatch Training using NumPy NPZ Archives
[TensorFlow 1: GitHub | Nbviewer]
Storing an Image Dataset for Minibatch Training using HDF5
[TensorFlow 1: GitHub | Nbviewer]
Using Input Pipelines to Read Data from TFRecords Files
[TensorFlow 1: GitHub | Nbviewer]
Using Queue Runners to Feed Images Directly from Disk
[TensorFlow 1: GitHub | Nbviewer]
Using TensorFlow's Dataset API
[TensorFlow 1: GitHub | Nbviewer]

Training and Preprocessing

Saving and Loading Trained Models -- from TensorFlow Checkpoint Files and NumPy NPZ Archives
[TensorFlow 1: GitHub | Nbviewer]

—完—
为您推荐
“12306”的架构到底有多牛逼？
中国程序员34岁生日当天在美国遭抢笔记本电脑，追击歹徒被拖行后身亡，为什么会发生此类事件？
阿里如何抗住90秒100亿？看这篇你就明白了！
60个Chrome神器插件大收集：助你快速成为老司机，一键分析网站技术栈



    
深度学习必懂的13种概率分布