• Usage
• Credits
• Code
• Issues

Hands On Network Machine Learning with Python

This repository provides the programming component for an introductory book to network machine learning with python. Our book will be published with Cambridge University Press in early 2025. This book assumes a recent version of python (we use Python 3.11 and 3.12). The compiled figures for the book can be accessed at https://ericwb.me/textbook_figs. These figures were generated programmatically using a github workflow which follows the below-noted steps.

The book can be cited via the bib citation:

@book{ authors={Bridgeford, Eric W. and Loftus, Alexander R. and Vogelstein, Joshua T.}, title={Hands On Network Machine Learning with Python}, publisher={Cambridge University Press}, year={2025} }

and a citation compliant with MLA 9th edition is:

EW Bridgeford, AR Loftus, JT Vogelstein. Hands On Network Machine Learning with Python. Cambridge University Press, 2025.

Usage

Begin by cloning this repository locally:

git clone git@github.com:ebridge2/textbook_figs.git <destination>/<directory>/

Docker

The simplest way to interact with the book is via the docker container. You can obtain the docker container locally with a properly configured docker engine using:

docker pull neurodata/graph-stats-book

Opening notebooks for individual sections in a pre-configured jupyter notebook

This repository has been designed alongside the docker container to allow seamless usage with the book in conjunction with jupyter notebooks. To use the docker container to launch a jupyter server that you can access in your browser, use the following command:

docker run -ti -v <destination>/<directory>/textbook_figs:/home/book -p <port>:8888 neurodata/graph-stats-book \ jupyter-lab --ip=0.0.0.0 --port=8888 /home/book/ \ --NotebookApp.token="graphbook"

The jupyter server can then be accessed in your browser at localhost:<port>, with the password graphbook. You should then be able to navigate to an appropriate notebook to reproduce figures as-is in the textbook, or begin your own notebook for usage as you learn and experiment with the content of the book.

Compiling the figures via docker

To compile all of the figures in the manuscript, we recommend using the docker container. Enter the docker container while provisioning the repository, navigate to the appropriate directory, and then compiling the book:

docker run -ti --entrypoint /bin/bash -v <destination>/<directory>/textbook_figs:/home/book neurodata/graph-stats-book cd /book jupyter-book build textbook_figs/

A fully-rendered HTML version of the book will be built in <destination>/<directory>/textbook_figs/_build/html/index.html. You can open this with a web browser from your computer.

Compiling the figures locally

We would recommend setting up a virtual environment for the book, using virtualenv with a recent version of python3. Within this virtualenv, install the dependencies:

cd <destination>/<directory>/textbook_figs pip install -r requirements.txt

and then compile the book:

jupyter-book build textbook_figs/

Credits

This project is created using the excellent open source Jupyter Book project and the executablebooks/cookiecutter-jupyter-book template.

Code

Functions specific to this book - e.g., plotting functions we use regularly - have been stored in the corresponding package graphbook-code.

Issues

If you have any issues with reproduction or anything is unclear, please drop us an issue and we will get back to you as quickly as we can.