Changeset 14779


Ignore:
Timestamp:
03/23/17 10:25:16 (4 years ago)
Author:
gkronber
Message:

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

File:
1 edited

Legend:

Unmodified
Added
Removed
  • trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees/GradientBoostedTreesAlgorithm.cs

    r14523 r14779  
    255255          var classificationProblemData = new ClassificationProblemData(problemData.Dataset,
    256256            problemData.AllowedInputVariables, problemData.TargetVariable, problemData.Transformations);
    257           classificationModel.RecalculateModelParameters(classificationProblemData, classificationProblemData.TrainingIndices);
     257          classificationModel.SetThresholdsAndClassValues(new double[] { double.NegativeInfinity, 0.0 }, new []{ 0.0, 1.0 });
    258258
    259259          var classificationSolution = new DiscriminantFunctionClassificationSolution(classificationModel, classificationProblemData);
Note: See TracChangeset for help on using the changeset viewer.