Changeset 5681 for branches/DataAnalysis Refactoring/HeuristicLab.Problems.DataAnalysis/3.4

Timestamp:

03/15/11 08:25:27 (13 years ago)

Author:

gkronber

Message:

#1418 refactored threshold calculators.

Location:

branches/DataAnalysis Refactoring/HeuristicLab.Problems.DataAnalysis/3.4

Files:

• . (modified) (1 prop)
• DiscriminantFunctionClassificationModel.cs (modified) (1 diff)
• DiscriminantFunctionClassificationSolution.cs (modified) (2 diffs)
• HeuristicLab.Problems.DataAnalysis-3.4.csproj (modified) (2 diffs)
• Interfaces/Classification/IDiscriminantFunctionThresholdCalculator.cs (added)
• Tests (modified) (1 prop)
• ThresholdCalculators (added)
• ThresholdCalculators/AccuracyMaximizationThresholdCalculator.cs (added)
• ThresholdCalculators/NormalDistributionCutPointsThresholdCalculator.cs (added)
• ThresholdCalculators/ThresholdCalculator.cs (added)

branches/DataAnalysis Refactoring/HeuristicLab.Problems.DataAnalysis/3.4

@@ -3 +3 @@
 HeuristicLabProblemsDataAnalysisPlugin.cs
 obj
+*.vs10x

@@ -3 +3 @@
 HeuristicLabProblemsDataAnalysisPlugin.cs
 obj
+*.vs10x

branches/DataAnalysis Refactoring/HeuristicLab.Problems.DataAnalysis/3.4/DiscriminantFunctionClassificationModel.cs

@@ -64 +64 @@
 
     [StorableConstructor]
-    protected DiscriminantFunctionClassificationModel() : base() { }
+    protected DiscriminantFunctionClassificationModel(bool deserializing) : base(deserializing) { }
     protected DiscriminantFunctionClassificationModel(DiscriminantFunctionClassificationModel original, Cloner cloner)
       : base(original, cloner) {

branches/DataAnalysis Refactoring/HeuristicLab.Problems.DataAnalysis/3.4/DiscriminantFunctionClassificationSolution.cs

@@ -43 +43 @@
       : base(original, cloner) {
     }
-    public DiscriminantFunctionClassificationSolution(IRegressionModel model, IClassificationProblemData problemData)
-      : this(new DiscriminantFunctionClassificationModel(model, problemData.ClassValues, CalculateClassThresholds(model, problemData, problemData.TrainingIndizes)), problemData) {
+    public DiscriminantFunctionClassificationSolution(IRegressionModel model, IClassificationProblemData problemData, IEnumerable<double> classValues, IEnumerable<double> thresholds)
+      : this(new DiscriminantFunctionClassificationModel(model, classValues, thresholds), problemData) {
     }
     public DiscriminantFunctionClassificationSolution(IDiscriminantFunctionClassificationModel model, IClassificationProblemData problemData)
@@ -91 +91 @@
     }
     #endregion
-
-    private static double[] CalculateClassThresholds(IRegressionModel model, IClassificationProblemData problemData, IEnumerable<int> rows) {
-      double[] thresholds;
-      double[] classValues;
-      CalculateClassThresholds(problemData, model.GetEstimatedValues(problemData.Dataset, rows), problemData.Dataset.GetEnumeratedVariableValues(problemData.TargetVariable, rows), out classValues, out thresholds);
-      return thresholds;
-    }
-
-    public static void CalculateClassThresholds(IClassificationProblemData problemData, IEnumerable<double> estimatedValues, IEnumerable<double> targetClassValues, out double[] classValues, out double[] thresholds) {
-      int slices = 100;
-      List<double> estimatedValuesList = estimatedValues.ToList();
-      double maxEstimatedValue = estimatedValuesList.Max();
-      double minEstimatedValue = estimatedValuesList.Min();
-      double thresholdIncrement = (maxEstimatedValue - minEstimatedValue) / slices;
-      var estimatedAndTargetValuePairs =
-        estimatedValuesList.Zip(targetClassValues, (x, y) => new { EstimatedValue = x, TargetClassValue = y })
-        .OrderBy(x => x.EstimatedValue)
-        .ToList();
-
-      classValues = problemData.ClassValues.OrderBy(x => x).ToArray();
-      int nClasses = classValues.Length;
-      thresholds = new double[nClasses + 1];
-      thresholds[0] = double.NegativeInfinity;
-      thresholds[thresholds.Length - 1] = double.PositiveInfinity;
-
-      // incrementally calculate accuracy of all possible thresholds
-      int[,] confusionMatrix = new int[nClasses, nClasses];
-
-      // one threshold is always treated as binary separation of the remaining classes
-      for (int i = 1; i < thresholds.Length - 1; i++) {
-        double lowerThreshold = thresholds[i - 1];
-        double actualThreshold = Math.Max(lowerThreshold, minEstimatedValue);
-        double lowestBestThreshold = double.NaN;
-        double highestBestThreshold = double.NaN;
-        double bestClassificationScore = double.PositiveInfinity;
-        bool seriesOfEqualClassificationScores = false;
-
-        while (actualThreshold < maxEstimatedValue) {
-          double classificationScore = 0.0;
-
-          foreach (var pair in estimatedAndTargetValuePairs) {
-            //all positives
-            if (pair.TargetClassValue.IsAlmost(classValues[i - 1])) {
-              if (pair.EstimatedValue > lowerThreshold && pair.EstimatedValue < actualThreshold)
-                //true positive
-                classificationScore += problemData.GetClassificationPenalty(classValues[i - 1], classValues[i - 1]);
-              else
-                //false negative
-                classificationScore += problemData.GetClassificationPenalty(classValues[i], classValues[i - 1]);
-            }
-              //all negatives
-            else {
-              if (pair.EstimatedValue > lowerThreshold && pair.EstimatedValue < actualThreshold)
-                //false positive
-                classificationScore += problemData.GetClassificationPenalty(classValues[i - 1], classValues[i]);
-              else
-                //true negative, consider only upper class
-                classificationScore += problemData.GetClassificationPenalty(classValues[i], classValues[i]);
-            }
-          }
-
-          //new best classification score found
-          if (classificationScore < bestClassificationScore) {
-            bestClassificationScore = classificationScore;
-            lowestBestThreshold = actualThreshold;
-            highestBestThreshold = actualThreshold;
-            seriesOfEqualClassificationScores = true;
-          }
-            //equal classification scores => if seriesOfEqualClassifcationScores == true update highest threshold
-          else if (Math.Abs(classificationScore - bestClassificationScore) < double.Epsilon && seriesOfEqualClassificationScores)
-            highestBestThreshold = actualThreshold;
-          //worse classificatoin score found reset seriesOfEqualClassifcationScores
-          else seriesOfEqualClassificationScores = false;
-
-          actualThreshold += thresholdIncrement;
-        }
-        //scale lowest thresholds and highest found optimal threshold according to the misclassification matrix
-        double falseNegativePenalty = problemData.GetClassificationPenalty(classValues[i], classValues[i - 1]);
-        double falsePositivePenalty = problemData.GetClassificationPenalty(classValues[i - 1], classValues[i]);
-        thresholds[i] = (lowestBestThreshold * falsePositivePenalty + highestBestThreshold * falseNegativePenalty) / (falseNegativePenalty + falsePositivePenalty);
-      }
-    }
   }
 }

branches/DataAnalysis Refactoring/HeuristicLab.Problems.DataAnalysis/3.4/HeuristicLab.Problems.DataAnalysis-3.4.csproj

@@ -115 +115 @@
     <Compile Include="ClusteringSolution.cs" />
     <Compile Include="ClassificationEnsembleModel.cs" />
+    <Compile Include="Interfaces\Classification\IDiscriminantFunctionThresholdCalculator.cs" />
     <Compile Include="Interfaces\Classification\IClassificationEnsembleModel.cs" />
     <Compile Include="Interfaces\Regression\IRegressionEnsembleModel.cs" />
@@ -152 +153 @@
     <Compile Include="RegressionSolution.cs" />
     <Compile Include="TableFileParser.cs" />
+    <Compile Include="ThresholdCalculators\AccuracyMaximizationThresholdCalculator.cs" />
+    <Compile Include="ThresholdCalculators\NormalDistributionCutPointsThresholdCalculator.cs" />
+    <Compile Include="ThresholdCalculators\ThresholdCalculator.cs" />
     <None Include="HeuristicLab.snk" />
     <None Include="HeuristicLabProblemsDataAnalysisPlugin.cs.frame" />

branches/DataAnalysis Refactoring/HeuristicLab.Problems.DataAnalysis/3.4/Tests

@@ -1 +1 @@
 bin
 obj
+*.vs10x

@@ -1 +1 @@
 bin
 obj
+*.vs10x

@@ -3 +3 @@
 HeuristicLabProblemsDataAnalysisPlugin.cs
 obj
+*.vs10x

@@ -1 +1 @@
 bin
 obj
+*.vs10x