Hands-On Neural Networks with PyTorch 1.0: An all-in-one guide to implement deep learning architectures using Python

Learn the essential background and model functioning of various neural network architectures such as CNN. RNN. LSTM, GANs, Autoencoders and more using best practices from an industry expert About This Book Design, build, and optimize neural networks using a range of real-world datasets Integrate neural network architectures in your own applications using PyTorch Implement deep learning and artificial intelligence principles to build smart cognitive models Who This Book Is ForMachine learning practitioners, Deep learning and AI enthusiasts who are looking to get well versed with different neural network architectures using PyTorch 1.0 will find this book useful. This book will allow you to implement a range of neural networks using easy-to-follow examples. Working knowledge of Python programming language and machine learning basics are mandatory. What You Will Learn Learn data representation for Neural Networks to understand how models are built Understand the math behind every algorithm in Neural networks Delve into how data is initialized for a Neural network in PyTorch Get well versed with a probability distribution and random noise in GANs Explore the layers in CNN- Convolutions, Pooling, Fully Connected and more Develop an autoencoder architecture to generate images Get to grips with Tuning and Optimizing RL-Algorithms In DetailNeural networks are the next-generation techniques to build smart web applications, powerful image and speech recognition systems, and more. This book will be your handy guide to help you bring neural networks in your daily life using the PyTorch 1.0 offerings. The book will start with the basics and the required concepts to understand how neural network functions. We will cover simple-to-intermediate tasks by building neural networks using real-world datasets. We will use PyTorch to implement a range of neural networks - from the simple feedforward neural networks to multilayered perceptrons, and more. You will understand how to implement cutting-edge neural network architectures such as CNN, RNN, LSTM and more using varied examples. Later you will cover real-world use-cases such as Boltzmann machine, GANs, reinforcement learning to simplify your understanding of neural networks and their implementations. By the end of this book, you will be able to build smart AI applications or integrate neural networks into your existing applications.

About the Author :
Vihar Kurama is currently working as a machine learning engineer, specialized in building web, and deep learning algorithms. His areas of research include computer vision and neural networks. He has published three research papers where he worked on Image Denoising Techniques and Image Semantic Segmentation using deep neural networks. He regularly writes on medium relating to programming and artificial intelligence and talks over conferences and workshops. In his free time, he dabbles with psychology and computational neuroscience, trying to knot one on the other. He is also a voracious reader.