Muutke küpsiste eelistusi

Deep Learning with Python, Third Edition 3rd edition [Pehme köide]

4.57/5 (1424 hinnangut Goodreads-ist)
  • Formaat: Paperback / softback, 648 pages, kõrgus x laius x paksus: 255x187x36 mm, kaal: 1270 g, Illustrations
  • Ilmumisaeg: 22-Dec-2025
  • Kirjastus: Manning Publications
  • ISBN-10: 1633436586
  • ISBN-13: 9781633436589
  • Formaat: Paperback / softback, 648 pages, kõrgus x laius x paksus: 255x187x36 mm, kaal: 1270 g, Illustrations
  • Ilmumisaeg: 22-Dec-2025
  • Kirjastus: Manning Publications
  • ISBN-10: 1633436586
  • ISBN-13: 9781633436589
Why wait to master deep learning when the tools are already within reach? As AI reshapes every industry, learning how to build models like GPT or generate stunning images is no longer a luxuryit is a necessity. Many developers struggle with fragmented tutorials or outdated libraries, leaving them unsure how to translate theory into practice. What if you could gain hands-on experience with the latest tools and techniques, backed by expert guidance, and build models that deliver real results from the start? 





First-principles walkthroughs: Understand every layer, activation, and optimizer, so troubleshooting feels natural. 





Keras 3 showcase: Use the newest API features for faster, cleaner model pipelines. 





Multiframework primer: Compare TensorFlow, PyTorch, and JAX to pick the right tool every time. 





Generative AI chapters: Craft text with your own GPT-style model and create images using diffusion. 





Production guidance: Learn scaling, tuning, and deployment tips that move notebooks into real apps. 

Deep Learning with Python, Third Edition, by François Chollet and Matthew Watson, delivers an authoritative, code-first roadmap from the minds behind Keras. 

Each chapter builds knowledge step by step, pairing intuitive explanations with color-coded listings you can run immediately. Expanded coverage tackles transformers, diffusion, and hardware-friendly workflows while retaining the approachable tone that made previous editions bestsellers. 

By the final page you will confidently architect, train, and fine-tune state-of-the-art models, ready to solve vision, language, and forecasting problems in your own projects. 

Ideal for developers with intermediate Python skills who crave practical, future-proof AI expertise.
1 WHAT IS DEEP LEARNING?  

2 THE MATHEMATICAL BUILDING BLOCKS OF NEURAL NETWORKS 

3 INTRODUCTION TO TENSORFLOW, PYTORCH, JAX, AND KERAS 

4 CLASSIFICATION AND REGRESSION 

5 FUNDAMENTALS OF MACHINE LEARNING 

6 THE UNIVERSAL WORKFLOW OF MACHINE LEARNING 

7 A DEEP DIVE ON KERAS 

8 IMAGE CLASSIFICATION 

9 CONVNET ARCHITECTURE PATTERNS 

10 INTERPRETING WHAT CONVNETS LEARN 

11 IMAGE SEGMENTATION 

12 OBJECT DETECTION 

13 TIMESERIES FORECASTING 

14 TEXT CLASSIFICATION 

15 LANGUAGE MODELS AND THE TRANSFORMER 

16 TEXT GENERATION 

17 IMAGE GENERATION 

18 BEST PRACTICES FOR THE REAL WORLD 

19 THE FUTURE OF AI 

20 CONCLUSIONS
François Chollet is a software engineer at Google and creator of the Keras deep learning library. 

Matthew Watson is a core maintainer of the Keras deep learning library, focusing primarily on tools for Natural Language Processing.