Pro Deep Learning with TensorFlow 2.0

This book builds upon the foundations established in its first edition, with updated chapters and the latest code implementations to bring it up to date with Tensorflow 2.0. Pro Deep Learning with TensorFlow 2.0 begins with the mathematical and core technical foundations of deep learning. Next, you will learn about convolutional neural networks, including new convolutional methods such as dilated convolution, depth-wise separable convolution, and their implementation. You’ll then gain an understanding of natural language processing in advanced network architectures such as transformers and various attention mechanisms relevant to natural language processing and neural networks in general. As you progress through the book, you’ll explore unsupervised learning frameworks that reflect the current state of deep learning methods, such as autoencoders and variational autoencoders. The final chapter covers the advanced topic of generative adversarial networks and their variants, such as cycle consistency GANs and graph neural network techniques such as graph attention networks and GraphSAGE. Upon completing this book, you will understand the mathematical foundations and concepts of deep learning, and be able to use the prototypes demonstrated to build new deep learning applications. What You Will Learn Who This Book Is Data scientists and machine learning professionals, software developers, graduate students, and open source enthusiasts.