Pages
1250
Year
2025
Level
intermediate
Read time
32h
François Chollet, Matthew Watson · Manning Publications · 2025
Reviewed by Ashish Sheth · Updated May 2026
Deep Learning with Python
Now Covering Generative AI, Keras 3, PyTorch, and JAX
4.5 / 5
AMAZON · 25 RATINGS
deep learning · machine learning
SUBJECTS
What you'll come away with
01.
How neural networks actually compute, not just metaphors
02.
Keras 3 patterns that work across TensorFlow, PyTorch, and JAX
03.
When and why CNNs, transformers, and diffusion models are each the right choice
04.
How to build modern generative AI models in Keras
05.
How to debug a model that won't train
06.
The author's philosophical framing of intelligence and generalization
Strengths
+Written by the creator of Keras — authority is unmatched
+3rd edition (2025) adds JAX, PyTorch, generative AI, and Keras 3 multi-backend
+Clear, code-driven explanations without unnecessary math
+Develops genuine intuition, not just recipe-following
Caveats
−1,250 pages — significant time commitment
−Keras-first framing may feel indirect if you live in pure PyTorch
−Goes broad rather than deep on the newest LLM-era techniques (pair with AI Engineering for production LLM work)
★ 4.5 FROM 25 READERS ON AMAZON
Check Price on Amazon →
Read this if
→Developers comfortable with Python and basic ML who want deep learning fundamentals
→Engineers building vision, sequence, or generative models in production
→PyTorch users who want Keras 3 as a unified API across frameworks
Skip this if
—Engineers who only call LLM APIs and never train models
—Researchers wanting deep mathematical proofs
—Complete ML beginners (read Hands-On Machine Learning first)
Head-to-head comparisons
Deep Learning with Python vs Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow → Deep Learning with Python vs The Hundred-Page Machine Learning Book → Deep Learning with Python vs Generative Deep Learning → Deep Learning with Python vs Build a Large Language Model (From Scratch) → Deep Learning with Python vs Natural Language Processing with Transformers → Frequently asked
How is the 3rd edition different from the 2nd?
The 3rd edition (October 2025) adds Keras 3 multi-backend support, PyTorch and JAX primers, and full coverage of modern generative AI. It's also significantly longer (1,250 vs 504 pages). If you read the 2nd edition recently, the new content is the main reason to upgrade.
Should I read this if I only use PyTorch?
Yes. Keras 3 runs natively on PyTorch, JAX, and TensorFlow. The 3rd edition includes a PyTorch primer and most code examples work with the PyTorch backend without rewriting.
Is this book good for beginners?
It assumes Python and basic ML. If you've never touched ML, read Hands-On Machine Learning first. If you have, this is the cleanest path into modern deep learning.
Read this next
3 alternatives
Ready?
Check Price on Amazon →