Changeset 10568

Timestamp:

03/10/14 14:43:37 (10 years ago)

Author:

mkommend

Message:

#1998: Code cleanup in ZeroR classification algorithm.

File:

• branches/ClassificationModelComparison/HeuristicLab.Algorithms.DataAnalysis/3.4/Linear/ZeroR.cs (modified) (2 diffs)

branches/ClassificationModelComparison/HeuristicLab.Algorithms.DataAnalysis/3.4/Linear/ZeroR.cs

@@ -20 +20 @@
 #endregion
 
-using System.Collections.Generic;
 using System.Linq;
 using HeuristicLab.Common;
 using HeuristicLab.Core;
-using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
 using HeuristicLab.Optimization;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
 using HeuristicLab.Problems.DataAnalysis;
-using HeuristicLab.Problems.DataAnalysis.Symbolic;
-using HeuristicLab.Problems.DataAnalysis.Symbolic.Classification;
 
 namespace HeuristicLab.Algorithms.DataAnalysis {
@@ -62 +58 @@
       Dataset dataset = problemData.Dataset;
       string target = problemData.TargetVariable;
-      var classValuesEnumerator = problemData.ClassValues.GetEnumerator();
-      var classValuesInDatasetEnumerator = dataset.GetDoubleValues(target, problemData.TrainingIndices).GetEnumerator();
+      var targetValues = dataset.GetDoubleValues(target, problemData.TrainingIndices);
 
-      Dictionary<double, int> classValuesCount = new Dictionary<double, int>(problemData.ClassValues.Count());
+      var dominantClass = targetValues.GroupBy(x => x).ToDictionary(g => g.Key, g => g.Count())
+        .MaxItems(kvp => kvp.Value).Select(x => x.Key).First();
 
-      //initialize
-      while (classValuesEnumerator.MoveNext()) {
-        classValuesCount[classValuesEnumerator.Current] = 0;
-      }
-
-      //count occurence of classes
-      while (classValuesInDatasetEnumerator.MoveNext()) {
-        classValuesCount[classValuesInDatasetEnumerator.Current] += 1;
-      }
-
-      classValuesEnumerator.Reset();
-      double mostOccurences = -1;
-      double bestClass = double.NaN;
-      while (classValuesEnumerator.MoveNext()) {
-        if (classValuesCount[classValuesEnumerator.Current] > mostOccurences) {
-          mostOccurences = classValuesCount[classValuesEnumerator.Current];
-          bestClass = classValuesEnumerator.Current;
-        }
-      }
-
-      ConstantClassificationModel model = new ConstantClassificationModel(bestClass);
-      ConstantClassificationSolution solution = new ConstantClassificationSolution(model, (IClassificationProblemData)problemData.Clone());
-
+      var model = new ConstantClassificationModel(dominantClass);
+      var solution = new ConstantClassificationSolution(model, (IClassificationProblemData)problemData.Clone());
       return solution;
-    }
-
-    private static SymbolicDiscriminantFunctionClassificationModel CreateDiscriminantFunctionModel(ISymbolicExpressionTree tree,
-    ISymbolicDataAnalysisExpressionTreeInterpreter interpreter,
-    IClassificationProblemData problemData,
-    IEnumerable<int> rows,
-    IEnumerable<double> classValues) {
-      var model = new SymbolicDiscriminantFunctionClassificationModel(tree, interpreter, new AccuracyMaximizationThresholdCalculator());
-      IList<double> thresholds = new List<double>();
-      double last = 0;
-      foreach (double item in classValues) {
-        if (thresholds.Count == 0) {
-          thresholds.Add(double.NegativeInfinity);
-        } else {
-          thresholds.Add((last + item) / 2);
-        }
-        last = item;
-      }
-      model.SetThresholdsAndClassValues(thresholds, classValues);
-      return model;
     }
   }