Mastering PyTorch: Create and deploy deep learning models from CNNs to multimodal models, LLMs, and beyond
Language: English
Published by Packt Publishing, 2024
- Softcover
- Used
Seller: medimops, Berlin, Germanymedimops
AbeBooks seller since May 10, 2010
Condition: Used - Very good
US$ 42.63
Quantity: 1 available
Add to basketItem description from seller
Gut/Very good: Buch bzw. Schutzumschlag mit wenigen Gebrauchsspuren an Einband, Schutzumschlag oder Seiten. / Describes a book or dust jacket that does show some signs of wear on either the binding, dust jacket or pages.
Seller Inventory # M01801074305-V
- Title
- Mastering PyTorch: Create and deploy deep learning models from CNNs to multimodal models, LLMs, and beyond
- Author
- Ashish Ranjan Jha
- Publisher
- Packt Publishing
- Publication year
- 2024
- Condition
- very good
- Binding
- Soft cover
- Language
- English
- ISBN 10
- 1801074305
- ISBN 13
- 9781801074308
- Edition
- 2nd Edition
Master advanced techniques and algorithms for machine learning with PyTorch using real-world examples
Updated for PyTorch 2.x, including integration with Hugging Face, mobile deployment, diffusion models, and graph neural networks
Get With Your Book: PDF Copy, AI Assistant, and Next-Gen Reader Free
Key Features
- Understand how to use PyTorch to build advanced neural network models
- Get the best from PyTorch by working with Hugging Face, fastai, PyTorch Lightning, PyTorch Geometric, Flask, and Docker
- Unlock faster training with multiple GPUs and optimize model deployment using efficient inference frameworks
Book Description
PyTorch is making it easier than ever before for anyone to build deep learning applications. This PyTorch deep learning book will help you uncover expert techniques to get the most out of your data and build complex neural network models.
You’ll build convolutional neural networks for image classification and recurrent neural networks and transformers for sentiment analysis. As you advance, you'll apply deep learning across different domains, such as music, text, and image generation, using generative models, including diffusion models. You'll not only build and train your own deep reinforcement learning models in PyTorch but also learn to optimize model training using multiple CPUs, GPUs, and mixed-precision training. You’ll deploy PyTorch models to production, including mobile devices. Finally, you’ll discover the PyTorch ecosystem and its rich set of libraries. These libraries will add another set of tools to your deep learning toolbelt, teaching you how to use fastai to prototype models and PyTorch Lightning to train models. You’ll discover libraries for AutoML and explainable AI (XAI), create recommendation systems, and build language and vision transformers with Hugging Face.
By the end of this book, you'll be able to perform complex deep learning tasks using PyTorch to build smart artificial intelligence models.
What you will learn
- Implement text, vision, and music generation models using PyTorch
- Build a deep Q-network (DQN) model in PyTorch
- Deploy PyTorch models on mobile devices (Android and iOS)
- Become well versed in rapid prototyping using PyTorch with fastai
- Perform neural architecture search effectively using AutoML
- Easily interpret machine learning models using Captum
- Design ResNets, LSTMs, and graph neural networks (GNNs)
- Create language and vision transformer models using Hugging Face
Who this book is for
This deep learning with PyTorch book is for data scientists, machine learning engineers, machine learning researchers, and deep learning practitioners looking to implement advanced deep learning models using PyTorch. This book is ideal for those looking to switch from TensorFlow to PyTorch. Working knowledge of deep learning with Python is required.
Table of Contents
- Overview of Deep Learning using PyTorch
- Deep CNN architectures
- Combining CNNs and LSTMs
- Deep Recurrent Model Architectures
- Advanced Hybrid Models
- Graph Neural Networks
- Music and Text Generation with PyTorch
- Neural Style Transfer
- Deep Convolutional GANs
- Image Generation Using Diffusion
- Deep Reinforcement Learning
- Model Training Optimizations
- Operationalizing PyTorch Models into Production
- PyTorch on Mobile Devices
- Rapid Prototyping with PyTorch
- PyTorch and AutoML
- PyTorch and Explainable AI
- Recommendation Systems with TorchRec
- PyTorch and Hugging Face
"Synopsis" may belong to another edition of this title.
About the Author
Ashish Ranjan Jha received his bachelor's degree in electrical engineering from IIT Roorkee (India), a master's degree in Computer Science from EPFL (Switzerland), and an MBA degree from Quantic School of Business (Washington). He has received a distinction in all 3 of his degrees. He has worked for large technology companies, including Oracle and Sony as well as the more recent tech unicorns such as Revolut, mostly focused on artificial intelligence. He currently works as a machine learning engineer. Ashish has worked on a range of products and projects, from developing an app that uses sensor data to predict the mode of transport to detecting fraud in car damage insurance claims. Besides being an author, machine learning engineer, and data scientist, he also blogs frequently on his personal blog site about the latest research and engineering topics around machine learning.
"About the title" may belong to another edition of this title.
Shipping rates from Germany to U.S.A.
| Item | 10 to 20 business days | 10 to 20 business days |
|---|---|---|
| First item | US$ 11.54 | US$ 23.07 |
Payment methods
Store description
**WILLKOMMEN** im AbeBooks Shop von medimops - dein Onlineshop mit der größten Auswahl an Second Hand Büchern. Stöber dich durch über 4 Millionen Artikel und spare bis zu 70 % im Vergleich zum Neupreis. Wir prüfen alle Artikel sorgfältig auf Qualität und Funktionalität. **WELCOME** to the medimop's AbeBooks shop, the online shop with a large range of second-hand books. Browse more than 4 million articles and save up to 70% on the new price. All our articles are quality checked. ** BIENVENUE ** dans la boutique AbeBooks de momox shop! Spécialiste des articles culturels d'occasion, nous vous offrons une gamme diversifiée de livres d'occasion. Découvrez nos offres et économisez jusqu'à 70% du prix neuf! Nous grarantissons tous nos articles en bon état. ** BENVENUTI ** nello shop medimops di AbeBooks, lo shop online per i libri usati. Abbiamo piu´di 4 milioni di articoli e potete risparmiare fino al 70% rispetto al nuovo. Tutti i nostri prodotti sono controllati a mano per garantire la massima qualità. ** BIENVENIDO ** a la tienda de medimops en AbeBooks, la tienda online con una gran oferta de libros de segunda mano. Navegue por más de 4 millones de artículos y ahorre hasta un 70% sobre el precio nuevo. Todos nuestros artículos son de calidad comprobada. …
Seller's business information
momox SE
Schreiberhauer Straße 30
Berlin, Germany 10317
Terms of sale
1. Scope
For all orders via our store on the AbeBooks Marketplace, the following terms and conditions apply. Unless otherwise agreed, the inclusion of any terms and conditions of your own used by you is contradicted.
2. contracting party, conclusion of contract, correction options
The purchase contract is concluded with momox SE.
The subject of the contract is the sale of goods.
If an article is posted by us on AbeBooks, the activation of the offer page on AbeBooks is the binding offer to conclude a contract under the conditions contained in the article page.
The presentation of the products in the online store does not constitute a legally binding offer, but a non-binding online catalog. You have the option to complete the ordering process via the shopping cart system or the buy now function. You can initially place our products in the shopping cart without obligation and correct your entries at any time before submitting your binding order by using the correction tools provided and explained for this purpose in the order process. By clicking the order button you make a binding offer for the goods contained in the shopping cart. By pressing the button "Buy now" an order is automatically created, which will be sent to the delivery address you have provided and for which a payment is made with the specified payment method.
The confirmation of receipt of your order will be sent by e-mail immediately after sending the order. We accept your order within 2 days by us or AbeBooks on our behalf submitting/issuing a declaration of acceptance or shipping confirmation in a separate email or, if applicable, the payment transaction is carried out by the service provider or the selected payment service provider. The time of execution of the payment transaction depends on the respective selected payment method (see under "Payment"). The relevant alternative for you depends on which of the listed events occurs first.
3. contract language, contract text storage
The language(s) available for the conclusion of the contract: German
We store the text of the contract on our systems, which, however, are not accessible to you.
However, you have the possibility to view your last orders via your AbeBooks account. The full text of the contract is not accessible there.
4. delivery conditions
Shipping costs will be added to the indicated product prices. You can find out more about the amount of the shipping costs at the offers.
5. payment
The payment processing is done by AbeBooks. You can pay the invoice amount by credit card.
6. retention of title
The goods remain our property until full payment.
7. transport damage
If goods are delivered with obvious transport damage, please complain about such defects to the deliverer as soon as possible and contact us immediately. Failure to make a complaint or contact us has no consequences for your legal claims and their enforcement, especially your warranty rights. However, they help us to be able to assert our own claims against the carrier or the transport insurance.
8. warranty
The statutory liability for defects shall apply.
Customer service: Phone number: 0049 (0) 30/488288200
9 a.m. to 6 p.m., e-mail: service@momox-shop.uk
9. liability
For claims based on damage caused by us, our legal representatives or vicarious agents, we shall always be liable without limitation - in the event of injury to life, limb or health - in the event of intentional or grossly negligent breach of duty - in the event of warranty promises, insofar as agreed, or - insofar as the scope of application of the Product Liability Act is opened. In the event of a breach of essential contractual obligations, the fulfillment of which is a prerequisite for the proper execution of the contract and the observance of which the contractual partner may regularly rely on (cardinal obligations) due to slight negligence on our part, on the part of our legal representatives or vicarious agents, liability shall be limited to the amount of the damage foreseeable at the time of the conclusion of the contract, the occurrence of which must typically be expected. Otherwise, claims for damages are excluded.
10. dispute resolution
The European Commission provides a platform for online dispute resolution (ODR), which can be found here: https://ec.europa.eu/consumers/odr/. Consumers have the possibility to use this platform for the settlement of their disputes.
To settle disputes arising from a contractual relationship with a consumer or whether such a contractual relationship exists at all, we are obliged to participate in dispute resolution proceedings before a consumer arbitration board. The responsible body is the Universalschlichtungsstelle des Bundes am Zentrum für Schlichtung e.V., Straßburger Straße 8, 77694 Kehl am Rhein, www.verbraucher-schlichter.de. We will participate in a dispute resolution procedure before this body.
11. protection of minors
If your order includes goods whose sale is subject to age restrictions, we ensure that the customer has reached the required minimum age by using a reliable procedure involving a personal identity and age check. The delivery person will hand over the goods only after the age check has been carried out and only to the customer in person.
Delivery to packing stations is not possible in this case.