Book Image

Hands-On Generative Adversarial Networks with PyTorch 1.x

By : John Hany, Greg Walters
Book Image

Hands-On Generative Adversarial Networks with PyTorch 1.x

By: John Hany, Greg Walters

Overview of this book

With continuously evolving research and development, Generative Adversarial Networks (GANs) are the next big thing in the field of deep learning. This book highlights the key improvements in GANs over generative models and guides in making the best out of GANs with the help of hands-on examples. This book starts by taking you through the core concepts necessary to understand how each component of a GAN model works. You'll build your first GAN model to understand how generator and discriminator networks function. As you advance, you'll delve into a range of examples and datasets to build a variety of GAN networks using PyTorch functionalities and services, and become well-versed with architectures, training strategies, and evaluation methods for image generation, translation, and restoration. You'll even learn how to apply GAN models to solve problems in areas such as computer vision, multimedia, 3D models, and natural language processing (NLP). The book covers how to overcome the challenges faced while building generative models from scratch. Finally, you'll also discover how to train your GAN models to generate adversarial examples to attack other CNN and GAN models. By the end of this book, you will have learned how to build, train, and optimize next-generation GAN models and use them to solve a variety of real-world problems.

Related Content you might be interested in

Current Title:

Small book image
Hands-On Generative Adversarial Networks with PyTorch 1.x
John Hany | Greg Walters
Dec, 2019 | 312 pages
Small book image
Generative Adversarial Networks Projects
Kailash Ahirwar
Jan, 2019 | 316 pages
Small book image
Hands-On Generative Adversarial Networks with Keras
Rafael Valle
May, 2019 | 272 pages
Small book image
Hands-On Image Generation with TensorFlow
Soon Yau Cheong
Dec, 2020 | 306 pages
Table of Contents (15 chapters)
Preface
Preface
Who this book is for
What this book covers
To get the most out of this book
Get in touch
Free Chapter
1
Section 1: Introduction to GANs and PyTorch
Section 1: Introduction to GANs and PyTorch
2
Generative Adversarial Networks Fundamentals
Generative Adversarial Networks Fundamentals
Fundamentals of machine learning
Generator and discriminator networks
What GAN we do?
Summary
References and useful reading list
3
Getting Started with PyTorch 1.3
Getting Started with PyTorch 1.3
What's new in PyTorch 1.3?
CUDA – GPU acceleration for fast training and evaluation
Installing PyTorch on Windows and Linux
References and useful reading list
Summary
4
Best Practices for Model Design and Training
Best Practices for Model Design and Training
Model design cheat sheet
Model training cheat sheet
Efficient coding in Python
Summary
5
Section 2: Typical GAN Models for Image Synthesis
Section 2: Typical GAN Models for Image Synthesis
6
Building Your First GAN with PyTorch
Building Your First GAN with PyTorch
Introduction to Deep Convolutional GANs
Creating a DCGAN with PyTorch
Model training and evaluation
Moving to larger datasets
Having fun with the generator network
Summary
References and useful reading list
7
Generating Images Based on Label Information
Generating Images Based on Label Information
CGANs – how are labels used?
Generating images from labels with the CGAN
Working with Fashion-MNIST
InfoGAN – unsupervised attribute extraction
References and useful reading list
Summary
8
Image-to-Image Translation and Its Applications
Image-to-Image Translation and Its Applications
Using pixel-wise labels to translate images with pix2pix
Pix2pixHD – high-resolution image translation
CycleGAN – image-to-image translation from unpaired collections
Summary
Furthering reading
9
Image Restoration with GANs
Image Restoration with GANs
Image super-resolution with SRGAN
Generative image inpainting
Summary
Useful reading list and references
10
Training Your GANs to Break Different Models
Training Your GANs to Break Different Models
Adversarial examples – attacking deep learning models
Generative adversarial examples
Summary
References and further reading list
11
Image Generation from Description Text
Image Generation from Description Text
Text-to-image synthesis with GANs
Generating photo-realistic images with StackGAN++
Summary
Further reading
12
Sequence Synthesis with GANs
Sequence Synthesis with GANs
Text generation via SeqGAN – teaching GANs how to tell jokes
Speech quality enhancement with SEGAN
Summary
Further reading
13
Reconstructing 3D models with GANs
Reconstructing 3D models with GANs
Fundamental concepts in computer graphics
Designing GANs for 3D data synthesis
Summary
Further reading
14
Other Books You May Enjoy
Other Books You May Enjoy
Leave a review - let other readers know what you think

CycleGAN – image-to-image translation from unpaired collections

You may have noticed that, when training pix2pix, we need to determine a direction (AtoB or BtoA) that the images are translated to. Does this mean that, if we want to freely translate from image set A to image set B and vice versa, we need to train two models separately? Not with CycleGAN, we say!

CycleGAN was proposed by Jun-Yan Zhu, Taesung Park, and Phillip Isola, et. al. in their paper, Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks. It is a bidirectional generative model based on unpaired image collections. The core idea of CycleGAN is built on the assumption of cycle consistency, which means that if we have two generative models, G and F, that translate between two sets of images, X and Y, in which Y=G(X) and X=F(Y), we can naturally assume that F(G(X)) should be very...

Previous Section
End of Section 4
Next Section