Skip to content


Repository files navigation

Elegant SciPy

Try the live notebooks via MyBinder!

This is the online repository for the book Elegant SciPy, written by Juan Nunez-Iglesias (@jni), Harriet Dashnow (@hdashnow), and Stéfan van der Walt (@stefanv), and published by O'Reilly Media:

Elegant SciPy Cover

Using this book.

The text of the book (inside the markdown folder) is available under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Public License (see

All code is available under the BSD 3-clause license (see This includes:

  • Python code embedded in the text of the book.
  • Python code inside scripts/.
  • Our Makefile and .yml files.

The authors also encourage educators to use this book in their own classrooms for noncommercial instructional uses (i.e. for slide presentations in a university lecture), provided that there is proper attribution to the O’Reilly edition in each instance.

If you are unsure whether your use falls outside fair use or the permissions given above, contact us at

Pre-built Jupyter notebooks

You can find the pre-compiled Jupyter notebooks corresponding to the latest version of this repo at:

This includes the necessary data in the data folder.

Or, if you are feeling adventurous, you can compile the notebooks yourself by following the instructions below.

Building the IPython notebooks

This book was written in markdown, with notedown and jupyter nbconvert used to build the book. To recreate the book contents, install the dependencies, (see below), then run make all from this directory (assuming you are using Mac OS X or Linux).

For interactive exploration you probably don't want to pre-run all the code, but rather create the notebooks with pre-populated input cells. To do this, run, for example:

notedown --match python /markdown/ch5.markdown --output ch5.ipynb

to build a Jupyter notebook containing Chapter 5, which you can then step through by starting a notebook session in this directory:

jupyter notebook

Installing dependencies

First, we build an isolated environment as not to interrupt any existing setup you may have. This can be done using, e.g., Conda or Python's built-in virtual environment module:

Conda install

  1. Install conda or Anaconda

  2. Build an isolated environment called "elegant-scipy" and install the necessary dependencies:

conda env create --name elegant-scipy -f /path/to/elegant-scipy/environment.yml
  1. Activate the environment with conda activate elegant-scipy (or source activate elegant-scipy if using conda 4.3.x or earlier, or activate elegant-scipy on Windows)

Python venv module

  1. Create a new virtual environment:

    python -m venv --prompt elegant-scipy venv

    This will create a python virtual environment named elegant-scipy and store it in ./venv.

  2. Activate the newly-created environment:

    source venv/bin/activate
  3. Install the dependencies with pip

    pip install -r requirements.txt

You can remove the virtual environment at any time by deleting the venv/ directory.


To build the full book on Windows, you will at a minimum need the following additional software:

Building the book on Windows is likely to be challenging because we developed the build process on Mac and Linux. However, you should still be able to create conda environments (see "Installing dependencies", above), and run notedown, to make the Jupyter notebooks (see "Building the IPython notebooks", above).

If you encounter any problems, please raise an issue!

Building the complete book

We are using notedown to convert a markdown file to an IPython notebook, run it, and then convert to html. For ease of use, this is done using a Makefile.

You can use make to build all the chapters:

$ make all

Or to build just an individual chapter, specify the file you wish to create:

$ make html/ch1.html

To generate a zip file containing html of all chapters along with a table of contents (for easy sharing):

$ make zip

Building HTMLBook


  • npm: brew install npm (macOS) or sudo apt-get install npm (Ubuntu/Debian)
  • htmlbook: npm install -g htmlbook
  • blahtexml: brew install blahtexml or sudo apt-get install blahtexml

Then run make htmlbook.

The results can be checked in to

and built at

Building with Docker

  1. Install Docker
  2. Switch to the directory containing this file
  3. Run docker-compose up