Maximum absolute difference between old and new corresponding messages in a cluster graph, before calibration is completed
Calibrates cluster graph.
Calibrates cluster graph.
Progress monitoring. It is called by this method at the beginning of every iteration
Order of clusters in which messages are sent for a single iteration of Belief Propagation
Applies evidence and calibrates cluster graph.
Applies evidence and calibrates cluster graph.
Sequence of Tuple2[variableId, variable value]
Progress monitoring. It is called by this method at the beginning of every iteration
Log likelihood of evidence in a cluster graph.
Returns cluster belief.
Returns log likelihood of assignment for all variables in a cluster graph.
Returns log likelihood of assignment for all variables in a cluster graph.
Array of Tuple2[variableId, variable value]
Returns marginal factor for a variable in a cluster graph.
Returns marginal factor for a variable in a cluster graph.
Sets evidence in a cluster graph.
Sets evidence in a cluster graph. Once evidence is set in a cluster graph, it cannot be reverted.
Tuple2[variableId, variable value]
Maximum absolute difference between old and new corresponding messages in a cluster graph, before calibration is completed
Loopy BP calibration and inference in a cluster graph, presented in 'Daphne Koller, Nir Friedman. Probabilistic Graphical Models, Principles and Techniques, 2009' book.
Maximum absolute difference between old and new corresponding messages in a cluster graph, before calibration is completed