دانلود کتاب Deep learning with PyTorch: a practical approach to building neural network models using PyTorch

فهرست مطالب :

Contents......Page 5
Preface......Page 12
Chapter 1: Getting Started with Deep Learning Using PyTorch......Page 17
The history of AI......Page 18
Machine learning......Page 19
Deep learning......Page 20
Applications of deep learning......Page 21
Hype associated with deep learning ......Page 23
Hardware availability......Page 24
Deep learning frameworks......Page 26
PyTorch......Page 27
Summary......Page 28
Installing PyTorch......Page 29
Our first neural network......Page 30
Data preparation......Page 31
Matrix (2-D tensors)......Page 32
3-D tensors......Page 33
Slicing tensors ......Page 34
5-D tensors......Page 37
Tensors on GPU......Page 38
Variables......Page 39
Creating learnable parameters......Page 41
Neural network model......Page 42
Network implementation ......Page 43
Optimize the neural network ......Page 44
Dataset class......Page 45
DataLoader class......Page 46
Summary ......Page 47
Deep dive into the building blocks of neural networks......Page 48
Layers – fundamental blocks of neural networks......Page 50
Sigmoid......Page 52
ReLU......Page 54
PyTorch non-linear activations......Page 56
Model architecture for different machine learning problems......Page 57
Loss functions......Page 58
Optimizing network architecture......Page 60
Image classification using deep learning......Page 61
Loading data into PyTorch tensors......Page 65
Loading PyTorch tensors as batches......Page 66
Building the network architecture......Page 67
Training the model ......Page 69
Summary......Page 73
Three kinds of machine learning problems......Page 74
Unsupervised learning......Page 75
Machine learning glossary......Page 76
Evaluating machine learning models......Page 77
Training, validation, and test split......Page 78
Simple holdout validation......Page 79
K-fold validation with shuffling ......Page 80
Data redundancy......Page 81
Value normalization......Page 82
Feature engineering......Page 83
Overfitting and underfitting......Page 84
Reducing the size of the network......Page 85
Applying weight regularization......Page 86
Dropout......Page 87
Problem definition and dataset creation......Page 89
Measure of success ......Page 90
Baseline model......Page 91
Large model enough to overfit......Page 92
Applying regularization......Page 93
Learning rate picking strategies ......Page 94
Summary......Page 95
Chapter 5: Deep Learning for Computer Vision......Page 96
Introduction to neural networks......Page 97
MNIST – getting data......Page 98
Building a CNN model from scratch......Page 100
Conv2d......Page 102
Pooling......Page 106
View......Page 108
Training the model......Page 110
Classifying dogs and cats – CNN from scratch......Page 113
Classifying dogs and cats using transfer learning......Page 115
Creating and exploring a VGG16 model ......Page 117
Training the VGG16 model ......Page 119
Calculating pre-convoluted features......Page 122
Visualizing outputs from intermediate layers......Page 125
Summary......Page 130
Chapter 6: Deep Learning with Sequence Data and Text......Page 131
Working with text data......Page 132
Converting text into characters......Page 133
Converting text into words......Page 134
N-gram representation......Page 135
One-hot encoding......Page 136
Word embedding......Page 139
Training word embedding by building a sentiment classifier......Page 140
torchtext.data......Page 141
torchtext.datasets......Page 142
Building vocabulary......Page 143
Generate batches of vectors......Page 144
Creating a network model with embedding......Page 145
Training the model......Page 146
Downloading the embeddings......Page 148
Loading the embeddings in the model......Page 149
Freeze the embedding layer weights......Page 150
Recursive neural networks......Page 151
Understanding how RNN works with an example ......Page 152
LSTM networks......Page 155
Creating batches......Page 158
Creating the network ......Page 159
Training the model......Page 160
Convolutional network on sequence data......Page 161
Creating the network......Page 162
Training the model......Page 163
Summary......Page 164
Chapter 7: Generative Networks......Page 165
Neural style transfer ......Page 166
Loading the data......Page 168
Creating the VGG model......Page 170
Style loss......Page 171
Extracting the losses ......Page 174
Creating the optimizer......Page 177
Training......Page 178
Generative adversarial networks......Page 179
Defining the generator network......Page 181
Batch normalization......Page 182
Generator......Page 184
Defining the discriminator network......Page 186
Defining loss and optimizer......Page 187
Training the discriminator with fake images......Page 188
Training the complete network......Page 189
Inspecting the generated images......Page 191
Language modeling ......Page 192
Preparing the data ......Page 193
Batches ......Page 195
Defining a model based on LSTM......Page 196
Defining the train and evaluate functions......Page 198
Training the model......Page 201
Summary......Page 203
Modern network architectures......Page 204
ResNet ......Page 205
Creating PyTorch datasets......Page 208
Creating a ResNet model......Page 209
Extracting convolutional features......Page 210
Creating a simple linear model......Page 211
Training and validating the model......Page 212
Inception......Page 213
Extracting convolutional features using register_forward_hook......Page 217
Creating a new dataset for the convoluted features......Page 218
Training and validating the model......Page 219
Densely connected convolutional networks – DenseNet......Page 220
DenseBlock......Page 221
DenseLayer......Page 222
Creating a DenseNet model......Page 223
Creating a dataset and loaders......Page 224
Creating a fully connected model and train......Page 225
Model ensembling......Page 226
Extracting the image features......Page 228
Creating a custom dataset along with data loaders......Page 230
Training and validating the model......Page 231
Encoder-decoder architecture ......Page 233
Summary......Page 235
Overview......Page 236
Interesting ideas to explore......Page 237
Object detection......Page 238
Image segmentation......Page 239
fast.ai – making neural nets uncool again......Page 240
Summary......Page 241
Other Books You May Enjoy......Page 242
Index......Page 245