Changeset 14779 for trunk/sources/HeuristicLab.Algorithms.DataAnalysis

Timestamp:

03/23/17 10:25:16 (7 years ago)

Author:

gkronber

Message:

#2755: set discriminant function threshold to zero for GBT with logistic regression loss

File:

• trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees/GradientBoostedTreesAlgorithm.cs (modified) (1 diff)

trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees/GradientBoostedTreesAlgorithm.cs

@@ -255 +255 @@
           var classificationProblemData = new ClassificationProblemData(problemData.Dataset,
             problemData.AllowedInputVariables, problemData.TargetVariable, problemData.Transformations);
-          classificationModel.RecalculateModelParameters(classificationProblemData, classificationProblemData.TrainingIndices);
+          classificationModel.SetThresholdsAndClassValues(new double[] { double.NegativeInfinity, 0.0 }, new []{ 0.0, 1.0 });
 
           var classificationSolution = new DiscriminantFunctionClassificationSolution(classificationModel, classificationProblemData);