PyTorch Artificial Intelligence Fundamentals

Description

Artificial Intelligence (AI) continues to grow in popularity and disrupt a wide range of domains, but it is a complex and daunting topic. In this book, you'll get to grips with building deep learning apps, and how you can use PyTorch for research and solving real-world problems. This book uses a recipe-based approach, starting with the basics of tensor manipulation, before covering Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) in PyTorch. Once you are well-versed with these basic networks, you'll build a medical image classifier using deep learning. Next, you'll use TensorBoard for visualizations. You'll also delve into Generative Adversarial Networks (GANs) and Deep Reinforcement Learning (DRL) before finally deploying your models to production at scale. You'll discover solutions to common problems faced in machine learning, deep learning, and reinforcement learning. You'll learn to implement AI tasks and tackle real-world problems in computer vision, natural language processing (NLP), and other real-world domains. By the end of this book, you'll have the foundations of the most important and widely used techniques in AI using the PyTorch framework. Cover Title Page Copyright and Credits Dedication Contributors About Packt Table of Contents Preface Chapter 1: Working with Tensors Using PyTorch Technical requirements Installing PyTorch Creating tensors in PyTorch How to do it... How it works... There's more... See also Exploring the NumPy bridge How to do it... How it works... There's more... See also Exploring gradients How to do it... How it works... There's more... See also Viewing tensors in PyTorch How to do it... How it works... There's more... See also Chapter 2: Dealing with Neural Networks Technical requirements Defining the neural network class How to do it... How it works... There's more... See also Creating a fully connected network How to do it... How it works... There's more... See also Defining the loss function How to do it... How it works... There's more... See also Implementing optimizers How to do it... How it works... There's more... See also Implementing dropouts How to do it... How it works... There's more... See also Implementing functional APIs How to do it... How it works... There' s more... See also Chapter 3: Convolutional Neural Networks for Computer Vision Technical requirements Exploring convolutions How to do it... How it works... There's more... See also Exploring pooling How to do it... How it works... There's more... See also Exploring transforms How to do it... How it works... There's more... See also Performing data augmentation How to do it... How it works... There's more... See also Loading image data Getting ready How to do it... How it works... There's more... See also Defining the CNN architecture How to do it... How it works... There's more... See also Training an image classifier  How to do it... How it works... There's more... See also Chapter 4: Recurrent Neural Networks for NLP Introducing RNNs Technical requirements Tokenization How to do it... How it works... There's more... See also Creating fields How to do it... How it works... There's more... See also Developing a dataset Getting ready How to do it... How it works... There's more... See also Developing iterators How to do it... How it works... There's more... See also Exploring word embeddings How to do it... How it works... There's more... See also Building an LSTM network How to do it... How it works... There's more... See also Multilayer LSTMs How to do it... How it works... There's more... See also Bidirectional LSTMs Getting ready How to do it... How it works... There's more... See also Chapter 5: Transfer Learning and TensorBoard Technical requirements Adapting a pretrained model Getting ready How to do it... How it works... Implementing model training How to do it... How it works... Implementing model testing How to do it... How it works... Loading the dataset How to do it... How it works... Defining the TensorBoard writer Getting ready How to do it... How it works... Training the model and unfreezing layers How to do it... How it works... There's more... See also Chapter 6: Exploring Generative Adversarial Networks Technical requirements Creating a DCGAN generator How to do it... How it works... See also Creating a DCGAN discriminator Getting Ready  How to do it... How it works... See also Training a DCGAN model Getting Ready How to do it... How it works... There's more... See also Visualizing DCGAN results Getting Ready How to do it... How it works... There's more... See also Running PGGAN with PyTorch hub Getting ready How to do it... How it works... There's more... See also Chapter 7: Deep Reinforcement Learning Introducing deep RL Introducing OpenAI gym – CartPole Getting ready How to do it... How it works... There's more... See also Introducing DQNs How to do it... How it works... There's more... See also Implementing the DQN class Getting ready How to do it... How it works... There's more... See also Training DQN  How to do it... How it works... There's more... See also Introduction to Deep GA How to do it... How it works... There's more... See also Generating agents How to do it... How it works... See also Selecting agents How to do it... How it works... Mutating agents How to do it... How it works... Training Deep GA How to do it... How it works... There's more... See also Chapter 8: Productionizing AI Models in PyTorch Technical requirements Deploying models using Flask Getting ready How to do it... How it works... There's more... See also Creating a TorchScript How to do it... How it works... There's more...  See also Exporting to ONNX Getting ready How to do it... How it works... There's more... See also Other Books You May Enjoy Index