Book Image

Generative AI with Python and TensorFlow 2

By : Joseph Babcock, Raghav Bali
4 (1)
Book Image

Generative AI with Python and TensorFlow 2

4 (1)
By: Joseph Babcock, Raghav Bali

Overview of this book

Machines are excelling at creative human skills such as painting, writing, and composing music. Could you be more creative than generative AI? In this book, you’ll explore the evolution of generative models, from restricted Boltzmann machines and deep belief networks to VAEs and GANs. You’ll learn how to implement models yourself in TensorFlow and get to grips with the latest research on deep neural networks. There’s been an explosion in potential use cases for generative models. You’ll look at Open AI’s news generator, deepfakes, and training deep learning agents to navigate a simulated environment. Recreate the code that’s under the hood and uncover surprising links between text, image, and music generation.

Related Content you might be interested in

Current Title:

Small book image
Generative AI with Python and TensorFlow 2
Joseph Babcock | Raghav Bali
Apr, 2021 | 488 pages
Small book image
Hands-On Image Generation with TensorFlow
Soon Yau Cheong
Dec, 2020 | 306 pages
Small book image
Hands-On Generative Adversarial Networks with Keras
Rafael Valle
May, 2019 | 272 pages
Small book image
Deep Learning with TensorFlow 2 and Keras
Sujit Pal | Amita Kapoor | Antonio Gulli
Dec, 2019 | 646 pages
Table of Contents (16 chapters)
Preface
Preface
Who this book is for
What this book covers
To get the most out of this book
Get in touch
1
An Introduction to Generative AI: "Drawing" Data from Models
An Introduction to Generative AI: "Drawing" Data from Models
Applications of AI
The rules of probability
Why use generative models?
Style transfer and image transformation
Unique challenges of generative models
Summary
References
Free Chapter
2
Setting Up a TensorFlow Lab
Setting Up a TensorFlow Lab
Deep neural network development and TensorFlow
VSCode
Docker: A lightweight virtualization solution
Kubernetes: Robust management of multi-container applications
Kubeflow: an end-to-end machine learning lab
A brief tour of Kubeflow's components
Kubeflow pipelines
Using Kubeflow Katib to optimize model hyperparameters
Summary
References
3
Building Blocks of Deep Neural Networks
Building Blocks of Deep Neural Networks
Perceptrons – a brain in a function
Multi-layer perceptrons and backpropagation
Varieties of networks: Convolution and recursive
Networks for sequence data
Building a better optimizer
Summary
References
4
Teaching Networks to Generate Digits
Teaching Networks to Generate Digits
The MNIST database
Restricted Boltzmann Machines: generating pixels with statistical mechanics
Stacking Restricted Boltzmann Machines to generate images: the Deep Belief Network
Creating an RBM using the TensorFlow Keras layers API
Creating a DBN with the Keras Model API
Summary
References
5
Painting Pictures with Neural Networks Using VAEs
Painting Pictures with Neural Networks Using VAEs
Creating separable encodings of images
The variational objective
Inverse Autoregressive Flow
Importing CIFAR
Creating the network from TensorFlow 2
Summary
References
6
Image Generation with GANs
Image Generation with GANs
The taxonomy of generative models
Generative adversarial networks
Vanilla GAN
Improved GANs
Progressive GAN
Challenges
Summary
References
7
Style Transfer with GANs
Style Transfer with GANs
Paired style transfer using pix2pix GAN
Unpaired style transfer using CycleGAN
Related works
Summary
References
8
Deepfakes with GANs
Deepfakes with GANs
Deepfakes overview
Modes of operation
Key feature set
High-level workflow
Replacement using autoencoders
Re-enactment using pix2pix
Challenges
Off-the-shelf implementations
Summary
References
9
The Rise of Methods for Text Generation
The Rise of Methods for Text Generation
Representing text
Text generation and the magic of LSTMs
LSTM variants and convolutions for text
Summary
References
10
NLP 2.0: Using Transformers to Generate Text
NLP 2.0: Using Transformers to Generate Text
Attention
GPT 1, 2, 3…
Summary
References
11
Composing Music with Generative Models
Composing Music with Generative Models
Getting started with music generation
Music generation using LSTMs
Music generation using GANs
MuseGAN – polyphonic music generation
Summary
References
12
Play Video Games with Generative AI: GAIL
Play Video Games with Generative AI: GAIL
Reinforcement learning: Actions, agents, spaces, policies, and rewards
Running GAIL on PyBullet Gym
Summary
References
13
Emerging Applications in Generative AI
Emerging Applications in Generative AI
Finding new drugs with generative models
Solving partial differential equations with generative modeling
Few shot learning for creating videos from images
Generating recipes with deep learning
Summary
References
Why subscribe?
14
Other Books You May Enjoy
Other Books You May Enjoy
15
Index
Index

Re-enactment using pix2pix

Re-enactment is another mode of operation for the deepfakes setup. It is supposedly better at generating believable fake content compared to the replacement mode. In earlier sections, we discussed different techniques used to perform re-enactment, i.e. by focusing on gaze, expressions, the mouth, and so on.

We also discussed image-to-image translation architectures in Chapter 7, Style Transfer with GANs. Particularly, we discussed in detail how the pix2pix GAN is a powerful architecture which enables paired translation tasks. In this section, we will leverage the pix2pix GAN to develop a face re-enactment setup from scratch. We will work toward building a network where we can use our own face, mouth, and expressions to control Barack Obama's (former US president) face. We will go through each and every step, starting right from preparing the dataset, to defining the pix2pix architecture, to finally generating the output re-enactment. Let's...

Previous Section
End of Section 7
Next Section