Changeset 8030 for trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression

Timestamp:

06/18/12 10:55:52 (12 years ago)

Author:

gkronber

Message:

#1874: reverted r8026. The behaviour of calculating the scaling parameters based on the unbounded values is intentional.

File:

• trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SymbolicRegressionModel.cs (modified) (1 diff)

trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SymbolicRegressionModel.cs

@@ -75 +75 @@
       var rows = problemData.TrainingIndizes;
       var estimatedValues = model.Interpreter.GetSymbolicExpressionTreeValues(model.SymbolicExpressionTree, dataset, rows);
-      var boundedEstimatedValues = estimatedValues.LimitToRange(model.LowerEstimationLimit, model.UpperEstimationLimit);
       var targetValues = dataset.GetDoubleValues(targetVariable, rows);
       double alpha;
       double beta;
       OnlineCalculatorError errorState;
-      OnlineLinearScalingParameterCalculator.Calculate(boundedEstimatedValues, targetValues, out alpha, out beta, out errorState);
+      OnlineLinearScalingParameterCalculator.Calculate(estimatedValues, targetValues, out alpha, out beta, out errorState);
       if (errorState != OnlineCalculatorError.None) return;