Changeset 131

Timestamp:

06/28/16 11:06:26 (8 years ago)

Author:

sherbold

Message:

• added workaround to WekaBaseTraining? to allow setting that no ADTree is used for BayesNet? when using a Vote classifier (required due to bug in WEKAs option parser

File:

• trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaBaseTraining.java (modified) (2 diffs)

trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaBaseTraining.java

@@ -22 +22 @@
 import weka.core.OptionHandler;
 import weka.classifiers.Classifier;
+import weka.classifiers.bayes.BayesNet;
 import weka.classifiers.meta.CVParameterSelection;
+import weka.classifiers.meta.Vote;
 
 /**
@@ -93 +95 @@
             cl = obj;
 
+            if( cl instanceof Vote ) {
+                Vote votingClassifier = (Vote) cl;
+                for( Classifier classifier : votingClassifier.getClassifiers() ) {
+                    if( classifier instanceof BayesNet ) {
+                        ((BayesNet) classifier).setUseADTree(false);
+                    }
+                }
+            }
             // we have cross val params
             // cant check on cvparam.length here, it may not be initialized