The PILGRIM (ProbabIListic Graphical and RelatIonal Models) library aims at providing efficient software tools in order to define and learn different probabilistic graphical models (currently directed ones: Bayesian networks, Probabilistic Relational Models, Graphical Event Models), and use them to make inference on data. It is composed of 3 different projects:
- PILGRIM General allows to define, learn and serialize Bayesian networks (including Dynamic Bayesian networks). It offers structure learning algorithms for Bayesian networks (e.g. greedy search, MMHC), structure evaluation (e.g. BDeu, BIC, qNML, SHD).
- PILGRIM Relational offers many features for the definition and learning of Probabilistic Relational Models (PRM) and their extensions, as well as their instantiation into grounded models for inference.
- PILGRIM Evential allows to define, learn and serialize Graphical Event Models. It includes structure learning algorithm (greedy search) as well as approximate probabilistic inference algorithms.
PILGRIM is actively developed and maintained by members of the Data, User, Knowledge (DUKe) team of the CNRS LS2N Laboratory, in Nantes, France, with the help of Ouest Valorisation, Technology transfer acceleration company. The project is led by Pr. Philippe Leray.
ProBT dependency
PILGRIM General and Relational projects encapsulate the ProBT library, developed by ProbaYes. ProBT allows to model, learn and make inference in Bayesian networks and PILGRIM reuses many of its classes defining basic concepts in probabilistic programming, while extending its capabilities in terms of Bayesian networks structure learning and in terms of relational datasets support.