INDRA (Integrated Network and Dynamical Reasoning Assembler) is an automated model assembly system interfacing with NLP systems and databases to collect knowledge, and through a process of assembly, produce causal graphs and dynamical models.
There was a bug in #1295 where setting include_more_specific=False in CountsScorer led to an error in subsequent use of the scorer to when called with BeliefEngine.set_hierarchy_probs. Where the statement feature matrix should have simply ignored any evidences from more specific statements it was instead expanding the feature matrix with additional columns which then did not match the trained classifier within the CountsScorer.
This PR adds a fix and a regression test for this error.
There was a bug in #1295 where setting
include_more_specific=Falsein CountsScorer led to an error in subsequent use of the scorer to when called withBeliefEngine.set_hierarchy_probs. Where the statement feature matrix should have simply ignored any evidences from more specific statements it was instead expanding the feature matrix with additional columns which then did not match the trained classifier within the CountsScorer.This PR adds a fix and a regression test for this error.