This is a Python implementation of the Dunn index, which is used to evaluate clustering results. While, officially, I followed the reference below, unnoficially, Wikipedia was a lot more helpful in this endeavour.
I also want to thank the people who used the code and commented in the GitHub gist where the code below is hosted. I received many bug reports and useful suggestions.
Kovács, F.; Legány, C. & Babos, A. Cluster Validity Measurement Techniques. 6th International Symposium of Hungarian Researchers on Computational Intelligence, 2005.