fchollet/deep-learning-with-python-notebooks

This repository contains the companion Jupyter notebooks for the book "Deep Learning with Python" by Francois Chollet (the creator of Keras) and Matthew Watson. Each notebook corresponds to a chapter in the book and contains the runnable code examples from that chapter, but without the explanatory text, figures, or pseudocode that appear in the printed book. The repository covers the third edition (2025), the second edition (2021), and the first edition (2017). The notebooks are written using Keras 3, a high-level deep learning library that can run on top of three different underlying frameworks: JAX, TensorFlow, or PyTorch. You choose which one to use by setting an environment variable at the top of the notebook. The code topics covered span the full range of modern deep learning: the mathematical foundations of neural networks, image classification, convolutional network architectures, image segmentation and object detection, time series forecasting, text classification, language models and the Transformer architecture (the foundation of modern AI systems like ChatGPT), text generation, and image generation. The easiest way to run the notebooks is Google Colab, a free browser-based environment that provides GPU access, the heavy computing power needed to train neural networks. No local installation is required. Some chapters use datasets from Kaggle, an online machine learning platform, which requires a free account. You would use this repository alongside the book if you are learning deep learning and want to run and experiment with the code examples as you read. The notebooks are not standalone tutorials without the book.