Cover; Title Page; Copyright and Credits; About Packt; Contributors; Table of Contents; Preface; Chapter 1: Introduction to PyTorch; What is PyTorch?; Installing PyTorch; Digital Ocean; Tunneling in to IPython; Amazon Web Services (AWS); Basic PyTorch operations; Default value initialization; Converting between tensors and NumPy arrays; Slicing and indexing and reshaping; In place operations; Loading data; PyTorch dataset loaders; Displaying an image; DataLoader; Creating a custom dataset; Transforms; ImageFolder; Concatenating datasets; Summary; Chapter 2: Deep Learning Fundamentals
Approaches to machine learningLearning tasks; Unsupervised learning; Clustering; Principle component analysis; Reinforcement learning; Supervised learning; Classification; Evaluating classifiers; Features; Handling text and categories; Models; Linear algebra review; Linear models; Gradient descent; Multiple features; The normal equation; Logistic regression; Nonlinear models; Artificial neural networks; The perceptron; Summary; Chapter 3: Computational Graphs and Linear Models; autograd; Computational graphs; Linear models; Linear regression in PyTorch; Saving models; Logistic regression
Activation functions in PyTorchMulti-class classification example; Summary; Chapter 4: Convolutional Networks; Hyper-parameters and multilayered networks; Benchmarking models; Convolutional networks; A single convolutional layer; Multiple kernels; Multiple convolutional layers; Pooling layers; Building a single-layer CNN; Building a multiple-layer CNN; Batch normalization; Summary; Chapter 5: Other NN Architectures; Introduction to recurrent networks; Recurrent artificial neurons ; Implementing a recurrent network; Long short-term memory networks; Implementing an LSTM
Building a language model with a gated recurrent unitSummary; Chapter 6: Getting the Most out of PyTorch; Multiprocessor and distributed environments; Using a GPU; Distributed environments; torch.distributed; torch.multiprocessing; Optimization techniques; Optimizer algorithms; Learning rate scheduler; Parameter groups; Pretrained models; Implementing a pretrained model; Summary; Other Books You May Enjoy; Index
Summary
PyTorch is extremely powerful and yet easy to learn. It provides advanced features such as supporting multiprocessor, distributed and parallel computation. This book is an excellent entry point for those wanting to explore deep learning with PyTorch to harness its power
