Changeset 7027

Timestamp:

11/21/11 16:31:44 (12 years ago)

Author:

gkronber

Message:

#1685: fixed a bug in the InteractiveSymbolicDiscriminantFunctionClassificationSolutionSimplifierView. The drawback is that the threshold calculation is much more complex now and can lead to an unresponsive UI as we do not calculate impacts in the background.

File:

• trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Classification.Views/3.4/InteractiveSymbolicDiscriminantFunctionClassificationSolutionSimplifierView.cs (modified) (1 diff)

trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Classification.Views/3.4/InteractiveSymbolicDiscriminantFunctionClassificationSolutionSimplifierView.cs

@@ -51 +51 @@
     protected override void UpdateModel(ISymbolicExpressionTree tree) {
       Content.Model = new SymbolicDiscriminantFunctionClassificationModel(tree, Content.Model.Interpreter);
-      // the default policy for setting thresholds in classification models is the accuarcy maximizing policy
-      // however for performance reasons we must use estimations of the normal distribution cut points as the thresholds
-      // here and in CalculateImpactValues as they are a lot faster to calculate
-      Content.SetClassDistibutionCutPointThresholds();
+      // the default policy for setting thresholds in classification models is the accuarcy maximizing policy.
+      // This is rather slow to calculate and can lead to a very laggy UI in the interactive solution simplifier.
+      // However, since we automatically prune sub-trees based on the threshold reaching the maximum accuracy we must
+      // also use maximum accuracy threshold calculation here in order to prevent incoherent behavior of the simplifier.
+      Content.SetAccuracyMaximizingThresholds();
     }