Network-Analysis-Made-Simple

Getting Set Up

For this tutorial, you will need the following packages:

Python 3
matplotlib
networkx
pandas
hiveplot - conda install -c ericmjl hiveplot or pip install hiveplot.
circos - conda install -c ericmjl circos or pip install circos
numpy
scipy

Clone the repository

$ cd /path/to/your/directory
Clone the repository to disk:
1. $ git clone https://github.com/ericmjl/Network-Analysis-Made-Simple.git
$ cd Network-Analysis-Made-Simple

Easiest way: Anaconda Distribution of Python

If you have the Anaconda distribution of Python 3 installed, then run bash conda-setup.sh, which wraps the commands below.

$ conda env create -f environment.yml
$ source activate network

Check your environment by running the following command:

$ python checkenv.py

If you do not have the Anaconda distribution, I would highly recommend getting it for Windows, Mac or Linux. It provides an isolated Python computing environment that will not interfere with your system Python installation, and comes with a very awesome package manager (conda) that makes installation of new packages a single conda install pkgname away.

If you're not using Python 3, then check out @jakevdp's talk at SciPy2015 to find out why!

Alternative to Anaconda: `pip install`

For those who do not have the capability of installing the Anaconda Python 3 distribution on their computers, please follow the instructions below.

Run bash venv-setup.sh, which wraps up the commands below. Special thanks to @matt-land for putting this script together.

Create a virtual environment for this tutorial, so that the installed packages do not mess with your regular Python environment. 2. $ pip install virtualenv 3. $ virtualenv network 4. $ source network/bin/activate
$ pip install matplotlib networkx pandas hiveplot numpy jupyter

Check your environment:

$ python checkenv.py

Run the Jupyter Notebook

$ jupyter notebook

Your browser will open to an index page where you can click on a notebook to run it.

Feedback

If you've attended this workshop, please leave feedback!.

Issues

Known Issues

If you get a "Python is not installed as a framework" error with matplotlib, please check out this issue for instructions to resolve it.

New Issues

If you're facing difficulties, please report it as an issue on this repository.

Credits

Divvy Data Challenge: https://www.divvybikes.com/datachallenge

Resources

Jon Charest's use of Circos plots to visualize networks of Metal music genres. blog post | notebook

People

Many thanks to the following users who have contributed to improving the content and finding bugs.

@jzab
@muthazhagu
@ghirlekar
@jamesbev
@romanchyla
@jonchar
@remik
@matt-land
@smacleod
@RyPeck
@zmilicc

willingc / network-analysis-made-simple Goto Github PK