Python implementation of exponential random graph models
DISCLAIMER: This package is still under development and woefully incomplete. I'd say "use at your own risk" but for now, it's really more like, "don't use." You have been warned!
An exponential random graph model (ergm) is a probability distribution over graphs specified by a set of sufficient statistics and and corresponding parameters.
The probability density function takes the form
Here, the k_a are the sufficient statistics (e.g. counts of edges, triangles, etc.) and the theta_a are the parameters.
Eventually, this package will support sampling and estimation of parameters from data.