Is deep-learning-with-python-notebooks open source?

Yes — fchollet/deep-learning-with-python-notebooks is open source, released under the MIT license.

What language is deep-learning-with-python-notebooks written in?

fchollet/deep-learning-with-python-notebooks is primarily written in Jupyter Notebook.

How popular is deep-learning-with-python-notebooks?

fchollet/deep-learning-with-python-notebooks has 20.2k stars on GitHub.

Where can I find deep-learning-with-python-notebooks?

fchollet/deep-learning-with-python-notebooks is on GitHub at https://github.com/fchollet/deep-learning-with-python-notebooks.

← all repositories

fchollet/deep-learning-with-python-notebooks

Keras 3 notebooks for a classic ML book, prose not included

Jupyter notebooks implementing the code samples from Deep Learning with Python, stripped of prose so you can execute instead of read.

★20.2k stars Jupyter Notebook Learning

View on GitHub ↗

Not currently ranked — collecting fresh signals.

star history

What it does This repository hosts the executable companions to François Chollet and Matthew Watson’s Deep Learning with Python (third edition, 2025), plus legacy notebooks for the 2017 and 2021 editions. Each notebook contains only runnable code blocks and section titles—no paragraphs, figures, or pseudocode—so you can step through the book’s examples directly. The collection spans the full arc of modern deep learning, from basic neural-network math to image generation and transformers.

The interesting bit The third-edition notebooks are built on Keras 3, which means you can swap JAX, TensorFlow, or PyTorch as the backend by setting an environment variable before importing Keras. That flexibility is unusual for a textbook companion repo, and it keeps the examples from feeling locked to any one framework ecosystem.

Key highlights

Covers chapters 2 through 18, including image segmentation, object detection, time-series forecasting, text generation, and image generation.
Designed to run on Colab’s free GPU tier, though later chapters will benefit from a faster GPU if you have one.
Includes legacy notebooks for the first and second editions of the book.
Some chapters rely on Kaggle datasets and require a Kaggle login to access weights and data.

Caveats

The notebooks deliberately omit explanatory text, so following along without the physical book is difficult.
Switching the Keras 3 backend mid-session requires restarting the runtime and re-running cells.
Chapters that use Kaggle data require authentication before those cells will run.

Verdict Grab this if you already own or are buying Deep Learning with Python and want to run the examples without transcribing them. If you are looking for a self-contained tutorial that explains the theory as it goes, this is explicitly not that.

Frequently asked

What is fchollet/deep-learning-with-python-notebooks?: Jupyter notebooks implementing the code samples from Deep Learning with Python, stripped of prose so you can execute instead of read.
Is deep-learning-with-python-notebooks open source?: Yes — fchollet/deep-learning-with-python-notebooks is open source, released under the MIT license.
What language is deep-learning-with-python-notebooks written in?: fchollet/deep-learning-with-python-notebooks is primarily written in Jupyter Notebook.
How popular is deep-learning-with-python-notebooks?: fchollet/deep-learning-with-python-notebooks has 20.2k stars on GitHub.
Where can I find deep-learning-with-python-notebooks?: fchollet/deep-learning-with-python-notebooks is on GitHub at https://github.com/fchollet/deep-learning-with-python-notebooks.