The Libra Toolkit

The Libra Toolkit is an open-source collection of algorithms for learning and inference with discrete probabilistic models such as Bayesian networks, Markov networks, dependency networks, sum-product networks, and arithmetic circuits. It is aimed at researchers, students, and practitioners in machine learning and probabilistic graphical models, with emphasis on structure learning, tractable models, and efficient exact inference. Libra is implemented as a set of command-line tools and is suitable for scripting and interactive use.