Python information theory computation. In most the computation, we use the true label/predicted label setting, i.e. the input labels are the true label and predicted label from a clustering algorithm or other methods.

• entropy
• normalized mutual information
• mutual information
• variation of information
• a toy implementation of Bayesian information criterion (BIC).

• Pedregosa, Fabian, et al. "Scikit-learn: Machine learning in Python." Journal of Machine Learning Research 12.Oct (2011): 2825-2830.