Hands-On Generative Adversarial Networks with PyTorch 2.x
Packt Publishing Limited (Verlag)
978-1-83508-438-0 (ISBN)
Key Features
Implement GAN architectures to generate images, text, audio, 3D models, and more
Understand how GANs work and become an active contributor in the open-source community
Learn about using GANs in combination with other generative models, such as Transformers and Diffusion Models
Book DescriptionGenerative AI is the most spoken of AI direction in media nowadays, and this book is aimed at assisting you in becoming an expert in its most well-established class of models - Generative Adversarial Nets.
With the help of this book, you will work your way up from understanding the basic components and architecture of GANs, building your first model from scratch to designing, building, training and optimizing a wide variety of these powerful models. You will go way beyond theoretical knowledge and gain hands-on experience in finding the right type of GAN for each specific problem using PyTorch examples provided in every chapter.
You will cover important image-generation and translation architectures such as classic and conditional GANs, DCGANs, StyleGANs, CycleGANs, and pix2pix. Learn to synthesize sequences, text and audio, and generate videos. Finally, we will dive into the state-of-the-art hybrid models of GANs with other generative models.
By the end of this book, you will be an expert in practical applications of GANs to real-world problems.What you will learn
Use PyTorch's latest features to ensure efficient model design
Get to grips with the working mechanisms of GAN models
Build and train a range of GANs to perform a variety of image synthesis and editing operations
Perform style transfer between unpaired and unpaired image collections with CycleGAN and pix2pix
Train GANs for video generation and video-to-video translation
Acquire the skills to use GANs for imitation learning and other automation tasks
Understand how to use elements of GANs in combination with other generative models
Attain a comprehensive understanding of the privacy and ethical considerations
Who this book is forThis GAN book is for machine learning practitioners and deep learning researchers looking to get hands-on guidance in implementing GAN models using PyTorch. You'll become familiar with state-of-the-art GAN architecture with the help of real-world examples. Working knowledge of Python programming language is necessary to grasp the concepts covered in this book.
Marija currently holds the position of a Senior Machine Learning Research Scientist at Metaphysic.ai. She has worked in Generative AI research for over 8 years, 5+ of them specifically with GANs. Previous employers and projects include Meta, iCAIRD (NHS Scotland), Seebyte (Batelle Company), and DREAM (EU Horizon2020). Her PhD is in Generative Models Applications to Robotics and Automation, acquired from the University of Edinburgh. She also has an MSc in Computational Statistics and Machine Learning from the University College London. She has authored multiple academic publications in several prestigious peer-reviewed venues in the fields of Machine Learning and Robotics, such as IEEE IROS, ICRA, and TPAMI.
Table of Contents
Basics of Generative Models
Getting Started with PyTorch 2.0
Useful tricks for model designing
Building your first GAN with PyTorch
Interactively generating images via Conditional GAN
Producing Photorealistic images with StyleGAN2/3
Image-to-image translation and its applications
Image restoration with GANs
Training your GANs to break other people’s models
Image generation from description text
Sequence synthesis with GANs
Video Generation with GANs
Reconstructing 3D models with GANs
GANs for Imitation Learning and Other Automation Tasks
Hybrid models: Using GANs in combination with Transformers or Diffusion Models
Where generative models are heading?
Commercial Use and Ethical Considerations when deploying generative models
| Erscheinungsdatum | 21.09.2023 |
|---|---|
| Verlagsort | Birmingham |
| Sprache | englisch |
| Maße | 191 x 235 mm |
| Themenwelt | Informatik ► Theorie / Studium ► Künstliche Intelligenz / Robotik |
| Informatik ► Weitere Themen ► Hardware | |
| ISBN-10 | 1-83508-438-9 / 1835084389 |
| ISBN-13 | 978-1-83508-438-0 / 9781835084380 |
| Zustand | Neuware |
| Haben Sie eine Frage zum Produkt? |
aus dem Bereich
