Book Image

Mastering PyTorch

By : Ashish Ranjan Jha
Book Image

Mastering PyTorch

By: Ashish Ranjan Jha

Overview of this book

Deep learning is driving the AI revolution, and PyTorch is making it easier than ever before for anyone to build deep learning applications. This PyTorch book will help you uncover expert techniques to get the most out of your data and build complex neural network models. The book starts with a quick overview of PyTorch and explores using convolutional neural network (CNN) architectures for image classification. You'll then work with recurrent neural network (RNN) architectures and transformers for sentiment analysis. As you advance, you'll apply deep learning across different domains, such as music, text, and image generation using generative models and explore the world of generative adversarial networks (GANs). You'll not only build and train your own deep reinforcement learning models in PyTorch but also deploy PyTorch models to production using expert tips and techniques. Finally, you'll get to grips with training large models efficiently in a distributed manner, searching neural architectures effectively with AutoML, and rapidly prototyping models using PyTorch and fast.ai. By the end of this PyTorch book, you'll be able to perform complex deep learning tasks using PyTorch to build smart artificial intelligence models.
Table of Contents (20 chapters)
1
Section 1: PyTorch Overview
4
Section 2: Working with Advanced Neural Network Architectures
8
Section 3: Generative Models and Deep Reinforcement Learning
13
Section 4: PyTorch in Production Systems

Chapter 11: Distributed Training

Before serving pre-trained machine learning models, which we discussed extensively in the previous chapter, we need to train our machine learning models. In Chapter 3, Deep CNN Architectures; Chapter 4, Deep Recurrent Model Architectures; and Chapter 5, Hybrid Advanced Models, we have seen the vast expanse of increasingly complex deep learning model architectures.

Such gigantic models often have millions and even billions of parameters. The recent (at the time of writing) Generative Pre-Trained Transformer 3 (GPT3) language model has 175 billion parameters. Using backpropagation to tune many parameters requires enormous amounts of memory and compute power. And even then, model training can take days to finish.

In this chapter, we will explore ways of speeding up the model training process by distributing the training task across machines and processes within machines. We will learn about the distributed training APIs offered by PyTorch –...