Official second-edition resources

PDFs, notebooks, and setup instructions

Get the official Machine Learning Refined downloads and choose how to run the Python notebooks. Start with a resource below, or browse chapters by topic. The PDF and slide indexes on GitHub link to the authors’ Dropbox downloads; notebooks and exercises live in the source repository.

Run the Python notebooks

Basic Python and matrix mathematics are the starting prerequisites. Choose Colab for a browser session, or local Jupyter to work with the repository on your computer.

Option 1: Open in Google Colab

  1. Choose an example in the visual notebook collection and follow its Colab link. You can also find available Colab links on the chapter pages.
  2. Sign in to Google if prompted. Save a copy in Drive if you want to keep your edits.
  3. Run cells from the top, including any setup or download cells, before running the examples. Read each notebook’s instructions for its data files and extra packages.

Most source notebooks can run in Colab without a local installation. Files and packages in a temporary runtime may need to be set up again when the session ends.

Option 2: Run locally with Jupyter

Install Git and Python 3.10 or newer first. Clone the full repository so the notebooks can find their helper modules and datasets. Run these commands in a terminal on macOS or Linux:

git clone https://github.com/neonwatty/machine-learning-refined.git
cd machine-learning-refined
python3 -m venv .venv
.venv/bin/python -m pip install -r requirements.txt
.venv/bin/python -m notebook --ip=127.0.0.1
Windows (PowerShell) commands
git clone https://github.com/neonwatty/machine-learning-refined.git
cd machine-learning-refined
py -3 -m venv .venv
.\.venv\Scripts\python.exe -m pip install -r requirements.txt
.\.venv\Scripts\python.exe -m notebook --ip=127.0.0.1

Open the local URL printed by Jupyter, including its sign-in token if requested. Browse to a notebook under notes/ or exercises/, select the environment’s Python kernel, and run its cells in order. Keep the terminal open while you work; stop the server with Ctrl+C when finished.

A missing module or data file usually means an environment or path needs attention. Check that you installed the repository’s requirements.txt in this environment and kept the repository folders intact. Some examples need additional tools such as FFmpeg for animation.

For alternative environments and notebook-specific guidance, consult the official installation instructions.

Before you download

Which edition do these resources accompany?

These collections accompany the second edition. The source notebooks include early-draft material; the final text expands on it and section numbering can differ. Use the chapter PDFs for the final presentation. The PDF and slide indexes label older first-edition collections separately.

Where are the exercises and datasets?

Start in the repository’s chapter-specific exercise folders. Keep the notebooks alongside the supplied data and helper directories; a single downloaded notebook may not contain everything it imports or reads. Check its instructions for any additional downloads.

What should I study first?

Start with the gradient descent guide for a compact NumPy example with a checked result, or use the book’s study roadmaps to plan a longer course.

Book identity

Machine Learning Refined: Foundations, Algorithms, and Applications

Second edition by Jeremy Watt, Reza Borhani, Aggelos K. Katsaggelos, published by Cambridge University Press on March 12, 2020.

Print ISBN
9781108480727
Digital ISBN
9781108690935