Changeset 7549 for branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/WeightCalculators/ContinuousPointCertaintyWeightCalculator.cs

Timestamp:

03/05/12 17:02:37 (12 years ago)

Author:

sforsten

Message:

#1776:

• models can be selected with a check box
• all strategies are now finished
• major changes have been made to provide the same behaviour when getting the estimated training or test values of an ensemble

File:

• branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/WeightCalculators/ContinuousPointCertaintyWeightCalculator.cs (modified) (3 diffs)

branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/WeightCalculators/ContinuousPointCertaintyWeightCalculator.cs

@@ -49 +49 @@
     }
 
-    protected override IEnumerable<double> DiscriminantCalculateWeights(ItemCollection<IDiscriminantFunctionClassificationSolution> discriminantSolutions) {
+    protected override IEnumerable<double> DiscriminantCalculateWeights(IEnumerable<IDiscriminantFunctionClassificationSolution> discriminantSolutions) {
       List<double> weights = new List<double>();
       IClassificationProblemData problemData = discriminantSolutions.ElementAt(0).ProblemData;
-      IEnumerable<double> targetValues = GetValues(problemData.Dataset.GetDoubleValues(problemData.TargetVariable).ToList(), problemData.TrainingIndizes);
+      IEnumerable<double> targetValues;
       IEnumerator<double> trainingValues;
 
       //only works for binary classification
       if (!problemData.ClassValues.Count().Equals(2))
-        return Enumerable.Repeat<double>(1, discriminantSolutions.Count);
+        return Enumerable.Repeat<double>(1, discriminantSolutions.Count());
 
       double maxClass = problemData.ClassValues.Max();
@@ -64 +64 @@
 
       foreach (var solution in discriminantSolutions) {
+        problemData = solution.ProblemData;
+        targetValues = GetValues(problemData.Dataset.GetDoubleValues(problemData.TargetVariable).ToList(), problemData.TrainingIndizes);
+        trainingValues = targetValues.GetEnumerator();
+
         IEnumerator<double> estimatedTrainingVal = solution.EstimatedTrainingValues.GetEnumerator();
         IEnumerator<double> estimatedTrainingClassVal = solution.EstimatedTrainingClassValues.GetEnumerator();
 
-        trainingValues = targetValues.GetEnumerator();
         double curWeight = 0.0;
         while (estimatedTrainingVal.MoveNext() && estimatedTrainingClassVal.MoveNext() && trainingValues.MoveNext()) {
-          //if (estimatedTrainingClassVal.Current.Equals(trainingValues.Current)) {
           if (trainingValues.Current.Equals(maxClass)) {
             if (estimatedTrainingVal.Current >= maxClass)
@@ -86 +88 @@
             }
           }
-          //}
         }
-        weights.Add(curWeight);
+        // normalize the weight (otherwise a model with a bigger training partition would probably be better)
+        weights.Add(curWeight / targetValues.Count());
       }
       return weights;