Generative Adversarial Networks Projects: Build next-generation generative models using TensorFlow and Keras

by Kailash Ahirwar

Explore various Generative Adversarial Network architectures using the Python ecosystem

Key FeaturesUse different datasets to build advanced projects in the Generative Adversarial Network domainImplement projects ranging from generating 3D shapes to a face aging applicationExplore the power of GANs to contribute in open source research and projectsBook DescriptionGenerative Adversarial Networks (GANs) have the potential to build next-generation models, as they can mimic any distribution of data. Major research and development work is being undertaken in this field since it is one of the rapidly growing areas of machine learning. This book will test unsupervised techniques for training neural networks as you build seven end-to-end projects in the GAN domain.

Generative Adversarial Network Projects begins by covering the concepts, tools, and libraries that you will use to build efficient projects. You will also use a variety of datasets for the different projects covered in the book. The level of complexity of the operations required increases with every chapter, helping you get to grips with using GANs. You will cover popular approaches such as 3D-GAN, DCGAN, StackGAN, and CycleGAN, and you’ll gain an understanding of the architecture and functioning of generative models through their practical implementation.

By the end of this book, you will be ready to build, train, and optimize your own end-to-end GAN models at work or in your own projects.

What you will learnTrain a network on the 3D ShapeNet dataset to generate realistic shapesGenerate anime characters using the Keras implementation of DCGANImplement an SRGAN network to generate high-resolution imagesTrain Age-cGAN on Wiki-Cropped images to improve face verificationUse Conditional GANs for image-to-image translationUnderstand the generator and discriminator implementations of StackGAN in KerasWho this book is forIf you’re a data scientist, machine learning developer, deep learning practitioner, or AI enthusiast looking for a project guide to test your knowledge and expertise in building real-world GANs models, this book is for you.

Table of ContentsIntroduction to Generative Adversarial Networks3D-GAN - Generating Shapes Using GANsFace Aging Using Conditional GANGenerating Anime Characters Using DCGANsUsing SRGANs to Generate Photo-Realistic ImagesStackGAN- Text to Photo-Realistic Image SynthesisCycleGAN- Turn Paintings into PhotosConditional GAN - Image-to-Image Translation Using Conditional Adversarial NetworksPredicting the Future of GANs

316 pages, Kindle Edition

Published January 31, 2019

Reviews

[Nguyen Le · Rating 1 out of 5]

I don't see the point of this book. If it was to show the theory behind, then the book just threw a bunch of concept together without explaining how they work, when to use them, etc. If it is to show how to code these projects, then the code level is very poor. A prime example is chapter 3, where the load_images function is obviously poorly constructed, leading to hours of waste time running that 1 function. A number of issue has been mentioned on github, but the author is unresponsive. It seems like the author himself has given up on this

[Niekoniecznie · Rating 2 out of 5]

If you're a beginner wanting to practice GANs with some projects, this is not the book for you. The code contains so many deprecated libraries, methods, and classes that it is difficult to correct yourself. The very first project contains a buggy function that makes your code run forever. The bug has been reported on the author's GitHub page, but they have not fixed it. I am giving 2 stars because the theory behind each network is explained well. I might come back to this book once I have more practice and want to understand the theory in more detail. Overall, I do not recommend this book.