Pedro Cuenca - Hands-On Generative AI with Transformers and Diffusion Models (First Early Release)
Here you can read online Pedro Cuenca - Hands-On Generative AI with Transformers and Diffusion Models (First Early Release) full text of the book (entire story) in english for free. Download pdf and epub, get meaning, cover and reviews about this ebook. year: 2023, publisher: OReilly Media, Inc., genre: Computer. Description of the work, (preface) as well as reviews are available. Best literature library LitArk.com created for fans of good reading and offers a wide selection of genres:
Romance novel
Science fiction
Adventure
Detective
Science
History
Home and family
Prose
Art
Politics
Computer
Non-fiction
Religion
Business
Children
Humor
Choose a favorite category and find really read worthwhile books. Enjoy immersion in the world of imagination, feel the emotions of the characters or learn something new for yourself, make an fascinating discovery.
- Book:Hands-On Generative AI with Transformers and Diffusion Models (First Early Release)
- Author:
- Publisher:OReilly Media, Inc.
- Genre:
- Year:2023
- Rating:3 / 5
- Favourites:Add to favourites
- Your mark:
Hands-On Generative AI with Transformers and Diffusion Models (First Early Release): summary, description and annotation
We offer to read an annotation, description, summary or preface (depends on what the author of the book "Hands-On Generative AI with Transformers and Diffusion Models (First Early Release)" wrote himself). If you haven't found the necessary information about the book — write in the comments, we will try to find it.
Pedro Cuenca: author's other books
Who wrote Hands-On Generative AI with Transformers and Diffusion Models (First Early Release)? Find out the surname, the name of the author of the book and a list of all author's works by series.
Hands-On Generative AI with Transformers and Diffusion Models (First Early Release) — read online for free the complete book (whole text) full work
Below is the text of the book, divided by pages. System saving the place of the last page read, allows you to conveniently read the book "Hands-On Generative AI with Transformers and Diffusion Models (First Early Release)" online for free, without having to search again every time where you left off. Put a bookmark, and you can go to the page where you finished reading at any time.
Font size:
Interval:
Bookmark:
by Pedro Cuenca , Apolinrio Passos , Omar Sanseviero , and Jonathan Whitaker
Copyright 2024 Pedro Cuenca, Apolinrio Passos, Omar Sanseviero, and Jonathan Whitaker. All rights reserved.
Printed in the United States of America.
Published by OReilly Media, Inc. , 1005 Gravenstein Highway North, Sebastopol, CA 95472.
OReilly books may be purchased for educational, business, or sales promotional use. Online editions are also available for most titles (http://oreilly.com). For more information, contact our corporate/institutional sales department: 800-998-9938 or corporate@oreilly.com.
- Acquisitions Editor: Nicole Butterfield
- Development Editor: Jill Leonard
- Production Editor: Gregory Hyman
- Interior Designer: David Futato
- Cover Designer: Karen Montgomery
- Illustrator: Kate Dullea
- September 2024: First Edition
- 2023-03-16: First Release
See http://oreilly.com/catalog/errata.csp?isbn=9781098149246 for release details.
The OReilly logo is a registered trademark of OReilly Media, Inc. Hands-On Generative AI with Transformers and Diffusion Models, the cover image, and related trade dress are trademarks of OReilly Media, Inc.
The views expressed in this work are those of the authors and do not represent the publishers views. While the publisher and the authors have used good faith efforts to ensure that the information and instructions contained in this work are accurate, the publisher and the authors disclaim all responsibility for errors or omissions, including without limitation responsibility for damages resulting from the use of or reliance on this work. Use of the information and instructions contained in this work is at your own risk. If any code samples or other technology this work contains or describes is subject to open source licenses or the intellectual property rights of others, it is your responsibility to ensure that your use thereof complies with such licenses and/or rights.
978-1-098-14924-6
With Early Release ebooks, you get books in their earliest formthe authors raw and unedited content as they writeso you can take advantage of these technologies long before the official release of these titles.
This will be the third chapter of the final book. Please note that the GitHub repo will be made active later on.
If you have comments about how we might improve the content and/or examples in this book, or if you notice missing material within this chapter, please reach out to the editor at .
In late 2020 a little-known class of models called diffusion modelsbegan causing a stir in the machine-learning world. Researchers figuredout how to use these models to generate synthetic images at higher qualitythan any produced by previous techniques. A flurry of papers followed,proposing improvements and modifications that pushed the quality up evenfurther. By late 2021 there were models like GLIDE that showcased incredibleresults on text-to-image tasks, and a few months later, these models hadentered the mainstream with tools like DALL-E 2 and Stable Diffusion.These models made it easy for anyone to generate images just by typingin a text description of what they wanted to see.
In this chapter, were going to dig into the details of how these modelswork. Well outline the key insights that make them so powerful,generate images with existing models to get a feel for how they work,and then train our own models to deepen this understanding further. The fieldis still rapidly evolving, but the topics covered here should give you asolid foundation to build on. Chapter 5 will explore more advanced techniques through the lens of a model called Stable Diffusion, and chapter 6 will explore applications of these techniques beyond simple image generation.
So what is it that makes diffusion models so powerful? Previoustechniques, such as VAEs or GANs, generate their final output via asingle forward pass of the model. This means the model must geteverything right on the first try. If it makes a mistake, it cant goback and fix it. Diffusion models, on the other hand, generate theiroutput by iterating over many steps. This iterative refinement allowsthe model to correct mistakes made in previous steps and graduallyimprove the output. To illustrate this, lets look at an example of adiffusion model in action.
We can load a pre-trained model using the Hugging Face diffuserslibrary. The pipeline can be used to create images directly, but thisdoesnt show us what is going on under the hood:
# Load the pipelineimage_pipe=DDPMPipeline.from_pretrained("google/ddpm-celebahq-256")image_pipe.to(device);# Sample an imageimage_pipe().images[0]
We can re-create the sampling process step by step to get a better lookat what is happening as the model generates images. We initialize oursample x with random noise and then run it through the model for 30steps. On the right, you can see the models prediction for what thefinal image will look like at specific steps - note that the initialpredictions are not particularly good! Instead of jumping right to thatfinal predicted image, we only modify x by a small amount in thedirection of the prediction (shown on the left). We then feed this new,slightly better x through the model again for the next step, hopefullyresulting in a slightly improved prediction, which can be used to updatex a little more, and so on. With enough steps, the model can producesome impressively realistic images.
# The random starting point for a batch of 4 imagesx=torch.randn(4,3,256,256).to(device)# Set the number of timesteps lowerimage_pipe.scheduler.set_timesteps(num_inference_steps=30)# Loop through the sampling timestepsfori,tinenumerate(image_pipe.scheduler.timesteps):# Get the prediction given the current sample x and the timestep twithtorch.no_grad():noise_pred=image_pipe.unet(x,t)["sample"]# Calculate what the updated sample should look like with the schedulerscheduler_output=image_pipeFont size:
Interval:
Bookmark:
Similar books «Hands-On Generative AI with Transformers and Diffusion Models (First Early Release)»
Look at similar books to Hands-On Generative AI with Transformers and Diffusion Models (First Early Release). We have selected literature similar in name and meaning in the hope of providing readers with more options to find new, interesting, not yet read works.
Discussion, reviews of the book Hands-On Generative AI with Transformers and Diffusion Models (First Early Release) and just readers' own opinions. Leave your comments, write what you think about the work, its meaning or the main characters. Specify what exactly you liked and what you didn't like, and why you think so.