Changeset 8814 for branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/WeightCalculators/ClassificationWeightCalculator.cs

Timestamp:

10/16/12 15:19:40 (12 years ago)

Author:

sforsten

Message:

#1776:

• improved performance of confidence calculation
• fixed bug in median confidence calculation
• fixed bug in average confidence calculation
• confidence calculation is now easier for training and test
• removed obsolete view ClassificationEnsembleSolutionConfidenceAccuracyDependence

File:

• branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/WeightCalculators/ClassificationWeightCalculator.cs (modified) (4 diffs)

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

@@ -24 +24 @@
 using HeuristicLab.Common;
 using HeuristicLab.Core;
-using HeuristicLab.Data;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
 
@@ -112 +111 @@
     }
 
-    public virtual double GetConfidence(IEnumerable<IClassificationSolution> solutions, int index, double estimatedClassValue) {
+    public virtual double GetConfidence(IEnumerable<IClassificationSolution> solutions, int index, double estimatedClassValue, CheckPoint handler) {
       if (solutions.Count() < 1)
         return double.NaN;
       Dataset dataset = solutions.First().ProblemData.Dataset;
       var correctSolutions = solutions.Select(s => new { Solution = s, Values = s.Model.GetEstimatedClassValues(dataset, Enumerable.Repeat(index, 1)).First() })
-                                      .Where(a => a.Values.Equals(estimatedClassValue))
+                                      .Where(a => handler(a.Solution.ProblemData, index) && a.Values.Equals(estimatedClassValue))
                                       .Select(a => a.Solution);
       return (from sol in correctSolutions
@@ -123 +122 @@
     }
 
-    public virtual IEnumerable<double> GetConfidence(IEnumerable<IClassificationSolution> solutions, IEnumerable<int> indices, IEnumerable<double> estimatedClassValue) {
+    public virtual IEnumerable<double> GetConfidence(IEnumerable<IClassificationSolution> solutions, IEnumerable<int> indices, IEnumerable<double> estimatedClassValue, CheckPoint handler) {
       if (solutions.Count() < 1)
         return Enumerable.Repeat(double.NaN, indices.Count());
 
+      List<int> indicesList = indices.ToList();
+
       Dataset dataset = solutions.First().ProblemData.Dataset;
-      Dictionary<IClassificationSolution, double[]> solValues = solutions.ToDictionary(x => x, x => x.Model.GetEstimatedClassValues(dataset, indices).ToArray());
+      Dictionary<IClassificationSolution, double[]> solValues = solutions.ToDictionary(x => x, x => x.Model.GetEstimatedClassValues(dataset, indicesList).ToArray());
       double[] estimatedClassValueArr = estimatedClassValue.ToArray();
-      double[] confidences = new double[indices.Count()];
+      double[] confidences = new double[indicesList.Count];
 
-      for (int i = 0; i < indices.Count(); i++) {
+      for (int i = 0; i < indicesList.Count; i++) {
         var correctSolutions = solValues.Where(x => DoubleExtensions.IsAlmost(x.Value[i], estimatedClassValueArr[i]));
         confidences[i] = (from sol in correctSolutions
+                          where handler(sol.Key.ProblemData, indicesList[i])
                           select weights[sol.Key]).Sum();
       }
@@ -142 +144 @@
 
     #region Helper
-    protected IEnumerable<double> GetValues(IList<double> targetValues, IEnumerable<int> indizes) {
-      return from i in indizes
-             select targetValues[i];
-    }
     protected bool PointInTraining(IClassificationProblemData problemData, int point) {
-      IntRange trainingPartition = problemData.TrainingPartition;
-      IntRange testPartition = problemData.TestPartition;
-      return (trainingPartition.Start <= point && point < trainingPartition.End)
-        && !(testPartition.Start <= point && point < testPartition.End);
+      return problemData.IsTrainingSample(point);
     }
     protected bool PointInTest(IClassificationProblemData problemData, int point) {
-      IntRange testPartition = problemData.TestPartition;
-      return testPartition.Start <= point && point < testPartition.End;
+      return problemData.IsTestSample(point);
     }
     protected bool AllPoints(IClassificationProblemData problemData, int point) {