Changeset 2136 for trunk/sources/HeuristicLab.SupportVectorMachines

Timestamp:

07/06/09 17:12:37 (16 years ago)

Author:

gkronber

Message:

Improved handling of exceptional cases in data-based modeling evaluators. #688 (SimpleEvaluators should handle exceptional cases more gracefully)

File:

• trunk/sources/HeuristicLab.SupportVectorMachines/3.2/VariableQualityImpactCalculator.cs (modified) (1 diff)

trunk/sources/HeuristicLab.SupportVectorMachines/3.2/VariableQualityImpactCalculator.cs

@@ -45 +45 @@
       for (int i = 0; i < end - start; i++) {
         values[i, 0] = SVM.Prediction.Predict(model.Model, scaledProblem.X[i]);
-        values[i, 1] = dataset.GetValue(start + i,targetVariable);
+        values[i, 1] = dataset.GetValue(start + i, targetVariable);
       }
 
-      return HeuristicLab.Modeling.SimpleMSEEvaluator.Calculate(values);
+      try { return HeuristicLab.Modeling.SimpleMSEEvaluator.Calculate(values); }
+      catch (ArgumentException) { return double.PositiveInfinity; }
     }
   }