Book Image

Generative Adversarial Networks Cookbook

By : Josh Kalin
Book Image

Generative Adversarial Networks Cookbook

By: Josh Kalin

Overview of this book

Developing Generative Adversarial Networks (GANs) is a complex task, and it is often hard to find code that is easy to understand. This book leads you through eight different examples of modern GAN implementations, including CycleGAN, simGAN, DCGAN, and 2D image to 3D model generation. Each chapter contains useful recipes to build on a common architecture in Python, TensorFlow and Keras to explore increasingly difficult GAN architectures in an easy-to-read format. The book starts by covering the different types of GAN architecture to help you understand how the model works. This book also contains intuitive recipes to help you work with use cases involving DCGAN, Pix2Pix, and so on. To understand these complex applications, you will take different real-world data sets and put them to use. By the end of this book, you will be equipped to deal with the challenges and issues that you may face while working with GAN models, thanks to easy-to-follow code solutions that you can implement right away.

Related Content you might be interested in

Current Title:

Small book image
Generative Adversarial Networks Cookbook
Josh Kalin
Dec, 2018 | 268 pages
Small book image
Generative Adversarial Networks Projects
Kailash Ahirwar
Jan, 2019 | 316 pages
Small book image
Keras Deep Learning Cookbook
Manpreet Singh Ghotra | Rajdeep Dua
Oct, 2018 | 252 pages
Small book image
Hands-On Generative Adversarial Networks with Keras
Rafael Valle
May, 2019 | 272 pages
Table of Contents (17 chapters)
Title Page
Title Page
Copyright and Credits
Copyright and Credits
About Packt
About Packt
Dedication
Dedication
Contributors
Contributors
Preface
Preface
Dedication2
Dedication2
Free Chapter
1
What Is a Generative Adversarial Network?
What Is a Generative Adversarial Network?
Introduction
Generative and discriminative models
A neural network love story
Deep neural networks
Architecture structure basics
Basic building block – generator
Basic building block – discriminator
Basic building block – loss functions
Training
GAN pieces come together in different ways
What does a GAN output?
Understanding the benefits of a GAN structure
Exercise
2
Data First, Easy Environment, and Data Prep
Data First, Easy Environment, and Data Prep
Introduction
Is data that important?
But first, set up your development environment
Data types
Data preprocessing
Anomalous data
Balancing data
Data augmentation
Exercise
3
My First GAN in Under 100 Lines
My First GAN in Under 100 Lines
Introduction
From theory to code – a simple example
Building a neural network in Keras and TensorFlow
Explaining your first GAN component – discriminator
Explaining your second GAN component – generator
Putting all the GAN pieces together
Training your first GAN
Training the model and understanding the GAN output
Exercise
4
Dreaming of New Outdoor Structures Using DCGAN
Dreaming of New Outdoor Structures Using DCGAN
Introduction
What is DCGAN? A simple pseudocode example
Tools – do I need any unique tools?
Parsing the data – is our data unique?
Code implementation – generator
Code implementation – discriminator
Training
Evaluation – how do we know it worked?
Adjusting parameters for better performance
Exercise
5
Pix2Pix Image-to-Image Translation
Pix2Pix Image-to-Image Translation
Introduction
Introducing Pix2Pix with pseudocode
Parsing our dataset
Code implementation – generator
Code – the GAN network
Code implementation – discriminator
Training
Exercise
6
Style Transfering Your Image Using CycleGAN
Style Transfering Your Image Using CycleGAN
Introduction
Pseudocode – how does it work?
Parsing the CycleGAN dataset
Code implementation – generator
Code implementation – discriminator
Code implementation – GAN
On to training
Exercise
7
Using Simulated Images To Create Photo-Realistic Eyeballs with SimGAN
Using Simulated Images To Create Photo-Realistic Eyeballs with SimGAN
Introduction
How SimGAN architecture works
Pseudocode – how does it work?
How to work with training data
Code implementation – loss functions
Code implementation – generator
Code implementation – discriminator
Code implementation – GAN
Training the simGAN network
Exercise
8
From Image to 3D Models Using GANs
From Image to 3D Models Using GANs
Introduction
Introduction to using GANs in order to produce 3D models
Environment preparation
Encoding 2D data and matching to 3D objects
Code implementation – generator
Code implementation – discriminator
Code implementation – GAN
Training this model
Exercise
Other Books You May Enjoy
Other Books You May Enjoy
Leave a review - let other readers know what you think
Index
Index

Generative and discriminative models

Machine learning (ML) and deep learning can be described by two terms: generative and discriminative modeling. When discussing the machine learning techniques that most people are familiar with, the thinking of a discriminative modeling technique, such as classification.

How to do it...

The difference between these two types of can be described by the following analogy:

  • Discriminative modeling: Observe paintings and determine the style of painting based on observations.

Here are a few steps that describe how we would do this in machine learning:

  1. First, we create a machine learning model that use convolutional layers or other learned features to understand the divisions in the data
  2. Next, we collect a dataset that has both a training set (60-90% of your data) and a validation dataset (10-40% of your data)
  3. Train the machine learning model using your data
  1.  Use this model to predict which datapoint belongs to a particular class - in our example, which painting belongs to which author
  • Generative modeling: Learn and reproduce paintings in various painters' styles and determine the painting style from the styles you learned.

Here are a few steps to describe a possible way to accomplish this type of modeling:

  1. Create a machine learning model that learns how to reproduce different painting styles
  2. Collect a training and validation dataset
  3. Train the machine learning model using the data
  4. Use this model to predict (inference) to produce examples of the paint author - use similarity metrics to verify the ability of the model to reproduce the painting style.

How it works...

Discriminative models will learn the boundary conditions between classes for a distribution:

  • Discriminative models get their power from more data
  • These models are not designed to work in an unsupervised manner or with unlabeled data

This can be described in a more graphical way, as follows:

  • Generative models will model the distribution of the classes for a given input distribution:
    • This creates a probabilistic model of each class in order to estimate the distribution
    • A generative model has the ability to use unlabeled data since it learns labels during the training process

This can be described in a more graphical way, as follows:

So, generative models are incredibly difficult to produce as they have to accurately model and reproduce the input distribution. The discriminative models are learning decision boundaries, which is why neural networks have been incredibly successful in recent years. The GAN architecture represents a radical departure from older techniques in the generative modeling area. We'll cover how neural networks are developed and then dive right in the GAN architecture development.

Previous Section
End of Section 3
Next Section