Graph modulus in Python

This repo contains a work-in-progress Python module for approximating the modulus for graphs introduced by Albin, Poggi-Corradini, and their colleagues.

Dependencies

Our code relies on the following modules:

• numpy for ...
• python-igraph for constructing and manipulating mathematical graphs
• cvxpy for solving convex optimization problems

See the file installation-notes.txt for more detail.

Organization

It is structured as a python module and can be installed on a shell as follows:

pip3 install ./pmodpy/

The main functions can be called as follows:

from pmodpy import modsubfamily
from pmodpy import modspans
from pmodpy import modwalks

If you would like to use the example graphs, include the following line:

from pmodpy.examplegraphs import examplegraphs

Then, one of the examples can be called as follow:

House=examplegraphs.House()

where House is the name of one of the example graphs.

If you want to compute the 2-modulus of walks from node 0 to node 1 on House graph with verbose option, then we do:

modwalks.modulus_walks_density(House, 0, 1, p=2, eps=2e-36, verbose=0)

##This graph is giving a different modulus as reported from the Shakeri paper.
Shakeri_1d =examplegraphs.Shakeri_1d()

Testing uses py.test

Acknowledgments

This is a limited adaptation of methods developed by the NODE Group at Kansas State University. Original work is licensed under the GNU Public License version 2.