pgmpy is a Python library for working with probabilistic graphical models (Bayesian Networks, Markov Network and variants). In the following months, I will extend pgmpy with methods to select Bayesian models based on data sets. I’ll first implement support for basic score-based and constraint-based structure learning. Second, I will add common enhancements to the score-based approach, such as local score computation/caching and tabu search. Finally, I will implement the MMHC algorithm, which combines the score-based and the constraint-based method.
This blog documents my progress in the scope of Google Summer of Code 2016. I will post updates on Bayesian Networks, structure learning and on implementation details.