<< Click to Display Table of Contents >> Navigation:  Using SMILE Wrappers > Inference > Most probable explanation

The Most probable explanation (MPE) algorithm in SMILE is a special case of the Annealed MAP algorithm and it solves the problem of finding the most likely configuration of values of all model nodes (those nodes that have been observed already have their most probable values) rather than a selected subset of nodes. The Annealed MAP algorithm is approximate and solves the problem by means of an approximate optimization procedure called simulated annealing.

MPE calculation takes into account any evidence currently set on nodes in the network.

Python

most_probable_explanation() -> pysmile.MpeResults

Java

MpeResults mostProbableExplanation();

C#

MpeResults mpeResults = MostProbableExplanation();

R

mpeResults <- mostProbableExplanation()

 

The MPE results object contains two integer arrays: one with the handles of unobserved nodes and another with the corresponding most probable outcomes for those nodes.