Deep Learning with Python was first published in 2017 by François Chollet. This third edition is freely available for anyone to read online via this site. If you would like to purchase a copy of the book, please visit Manning Press.
The book
This book was written for anyone who wishes to explore deep learning from scratch or broaden their understanding of deep learning. Whether you’re a practicing machine learning engineer, a software developer, or a college student, you’ll find value in these pages.
You’ll explore deep learning in an approachable way—starting simply and working up to state-of-the-art techniques. We hope you’ll find that this book strikes a balance between intuition, theory, and hands-on practice. It avoids mathematical notation, preferring instead to explain the core ideas of deep learning via functioning code paired with explanations of the underlying principles. You’ll train machine learning models from scratch in a number of different problem domains and learn practical recommendations for writing deep learning programs and deploying them in the real world.
After reading this book, you’ll have a solid understanding of what deep learning is, when it’s applicable, and what its limitations are. You’ll be familiar with the standard workflow for approaching and solving machine learning problems, and you’ll know how to address commonly encountered issues.
The reader
This book is written for people with some Python programming experience who want to get started with machine learning and deep learning. But this book can also be valuable to many different types of readers:
- If you’re a data scientist familiar with machine learning, this book will provide you with a solid, practical introduction to deep learning, the fastest-growing and most significant subfield of machine learning.
- If you’re a deep learning researcher or practitioner looking to get started with the Keras framework, you’ll find this book to be the ideal Keras crash course.
- If you’re a graduate student studying deep learning in a formal setting, you’ll find this book to be a practical complement to your education, helping you build intuition around the behavior of deep neural networks and familiarizing you with key best practices.
Even technically minded people who don’t code regularly will find this book useful as an introduction to both basic and advanced deep learning concepts.
To understand the code examples, you’ll need reasonable Python proficiency. You don’t need previous experience with machine learning or deep learning: this book covers, from scratch, all the necessary basics. You don’t need an advanced mathematics background either—high-school-level mathematics should suffice to follow along.
The code
This book is an example of literate programming—it is simultaneously a prose explanation of deep learning concepts and a runnable Python program. Starting with chapter 2, each chapter will be interspersed with code listings that can be run as a Jupyter notebook, an interactive notebook for running Python code and visualizing data. We strongly suggest running, modifying, and playing with the examples in this book as you read to build your own intuitions on the concepts presented.
All code is written in the Python language with the deep learning library Keras. Keras can be run on top of TensorFlow, PyTorch, and JAX, the most popular low-level deep learning frameworks as of 2025. All code can be run on a local machine or directly in the browser using Google Colab, a hosted environment for Jupyter notebooks.
On this web version, a Colab link will be available at the top of each chapter allowing you to run the chapter’s code in the browser on free, hosted hardware. All code is made available on GitHub, along with instructions for running the code locally.
The authors
François Chollet François Chollet has been working with deep learning since it started getting traction in academia in 2012. François is the author of Keras, one of the most popular libraries for deep learning. Keras is used in university classrooms; at companies like Google, Netflix, and Spotify; and in scientific organizations like CERN, NASA, and NIH. Francois is the co-founder of the Ndea research lab for frontier AI systems and created the ARC-AGI challenge for measuring machine intelligence.
Matthew Watson has been working on machine learning across Google since 2018, including the Gemini model and Google’s open source deep learning ecosystem. He is a core maintainer of Keras, focusing on Keras’s tools for natural language processing. He completed his master’s in computer science at Stanford University, researching procedural modeling techniques at the Stanford Graphics Lab.
Acknowledgements
First, we’d like to thank the Keras community for making this book possible. Over the past decade, Keras has grown to have thousands of open source contributors and more than 2 million users. Their contributions and feedback have turned Keras into what it is today.
On a personal note, François would like to thank his wife for her endless support during the development of Keras and the writing of this book. Matthew would like to thank his partner Kate, his family, and all the friends who have supported him along the way.
We thank the people at Manning who made this book possible: publisher Marjan Bace and everyone on the editorial and production teams, including Michael Stephens, Aleksandar Dragosavljević, and many others who worked behind the scenes. Many thanks go to the peer reviewers: Aakash Nain, Abheesht Sharma, Abhishek Shivanna, Aritra Roy Gosthipaty, Avinash Tiwari, Brandon Friar, Christopher Kardell, Srivathsan Srinivasagopalan, Edmon Begoli, Guillaume Alleon, Ian Stirk, Jacqueline Nolis, Kishore Reddy, Levi McClenny, Margaret Maynard-Reid, Nilson Chapagain, Prashanth Josyula, Preetish Kakkar, Sai Srinivas Somarouthu, Samuel Marks, Srivathsan Srinivasagopalan, Thiago Britto Borges, Todd Cook, and Varun Chawla -- and all the other people who sent us feedback on the draft of the book.
On the coding side, special thanks go to Tomasz Kalinowski, who contributed to code examples in this book; Ian Hough, who served as the book’s development editor; and Gabriel Rasskin, who served as the book’s technical proofreader.