Changeset 8921 for trunk

Timestamp:

11/19/12 15:11:44 (12 years ago)

Author:

gkronber

Message:

#1925 fixed robustness of NormalDistributionCutPointsThresholdCalculator. Now it always returns at least negative infinity and the most frequent class.

Location:

trunk/sources

Files:

• HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/SymbolicDiscriminantFunctionClassificationModel.cs (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/ThresholdCalculators/NormalDistributionCutPointsThresholdCalculator.cs (modified) (4 diffs)

trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/SymbolicDiscriminantFunctionClassificationModel.cs

@@ -88 +88 @@
       var classValuesArr = classValues.ToArray();
       var thresholdsArr = thresholds.ToArray();
-      if (thresholdsArr.Length != classValuesArr.Length) throw new ArgumentException();
+      if (thresholdsArr.Length != classValuesArr.Length || thresholdsArr.Length < 1) 
+        throw new ArgumentException();
+      if (!double.IsNegativeInfinity(thresholds.First())) 
+        throw new ArgumentException();
 
       this.classValues = classValuesArr;

trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/ThresholdCalculators/NormalDistributionCutPointsThresholdCalculator.cs

@@ -71 +71 @@
         }
       }
+
       double[] originalClasses = classMean.Keys.OrderBy(x => x).ToArray();
       int nClasses = originalClasses.Length;
@@ -98 +99 @@
       thresholdList.Add(double.PositiveInfinity);
 
-      // determine class values for each partition separated by a threshold by calculating the density of all class distributions
-      // all points in the partition are classified as the class with the maximal density in the parition
-      if (thresholdList.Count == 2) {
+
+      // find the most likely class for the points between thresholds m
+      List<double> filteredThresholds = new List<double>();
+      List<double> filteredClassValues = new List<double>();
+      for (int i = 0; i < thresholdList.Count - 1; i++) {
+        // determine class with maximal density mass between the thresholds
+        double maxDensity = DensityMass(thresholdList[i], thresholdList[i + 1], classMean[originalClasses[0]], classStdDev[originalClasses[0]]);
+        double maxDensityClassValue = originalClasses[0];
+        foreach (var classValue in originalClasses.Skip(1)) {
+          double density = DensityMass(thresholdList[i], thresholdList[i + 1], classMean[classValue], classStdDev[classValue]);
+          if (density > maxDensity) {
+            maxDensity = density;
+            maxDensityClassValue = classValue;
+          }
+        }
+        if (maxDensity > double.NegativeInfinity &&
+          (filteredClassValues.Count == 0 || !maxDensityClassValue.IsAlmost(filteredClassValues.Last()))) {
+          filteredThresholds.Add(thresholdList[i]);
+          filteredClassValues.Add(maxDensityClassValue);
+        }
+      }
+
+      if (filteredThresholds.Count == 0 || !double.IsNegativeInfinity(filteredThresholds.First())) {
         // this happens if there are no thresholds (distributions for all classes are exactly the same)
+        // or when the CDF up to the first threshold is zero
         // -> all samples should be classified as the class with the most observations
         // group observations by target class and select the class with largest count 
@@ -107 +129 @@
                               .OrderBy(g => g.Count())
                               .Last().Key;
-        thresholds = new double[] { double.NegativeInfinity };
-        classValues = new double[] { mostFrequentClass };
-      } else {
-
-        // at least one reasonable threshold ...
-        // find the most likely class for the points between thresholds m
-        List<double> filteredThresholds = new List<double>();
-        List<double> filteredClassValues = new List<double>();
-        for (int i = 0; i < thresholdList.Count - 1; i++) {
-          // determine class with maximal density mass between the thresholds
-          double maxDensity = DensityMass(thresholdList[i], thresholdList[i + 1], classMean[originalClasses[0]], classStdDev[originalClasses[0]]);
-          double maxDensityClassValue = originalClasses[0];
-          foreach (var classValue in originalClasses.Skip(1)) {
-            double density = DensityMass(thresholdList[i], thresholdList[i + 1], classMean[classValue], classStdDev[classValue]);
-            if (density > maxDensity) {
-              maxDensity = density;
-              maxDensityClassValue = classValue;
-            }
-          }
-          if (maxDensity > double.NegativeInfinity &&
-            (filteredClassValues.Count == 0 || !maxDensityClassValue.IsAlmost(filteredClassValues.Last()))) {
-            filteredThresholds.Add(thresholdList[i]);
-            filteredClassValues.Add(maxDensityClassValue);
-          }
-        }
-        thresholds = filteredThresholds.ToArray();
-        classValues = filteredClassValues.ToArray();
-      }
+        filteredThresholds.Insert(0, double.NegativeInfinity);
+        filteredClassValues.Insert(0, mostFrequentClass);
+      }
+
+      thresholds = filteredThresholds.ToArray();
+      classValues = filteredClassValues.ToArray();
     }
 
@@ -208 +208 @@
           double g = Math.Sqrt(2 * s2 * s2 * Math.Log(s2 / s1) - 2 * s1 * s1 * Math.Log(s2 / s1) - 2 * m1 * m2 + m1 * m1 + m2 * m2);
           double s = (s1 * s1 - s2 * s2);
-          x1 =  (m2 * s1 * s1 - m1 * s2 * s2 + s1 * s2 * g) / s;
+          x1 = (m2 * s1 * s1 - m1 * s2 * s2 + s1 * s2 * g) / s;
           x2 = -(m1 * s2 * s2 - m2 * s1 * s1 + s1 * s2 * g) / s;
         }