Changeset 12843 for branches/HiveStatistics/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4

Timestamp:

08/11/15 10:11:47 (9 years ago)

Author:

ascheibe

Message:

#2388 merged trunk into branch

Location:

branches/HiveStatistics/sources

Files:

• . (modified) (2 props)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Classification (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/Plugin.cs.frame (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/Properties/AssemblyInfo.cs.frame (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/SymbolicClassificationPruningAnalyzer.cs (modified) (2 diffs)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/SymbolicClassificationPruningOperator.cs (modified) (6 diffs)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/SymbolicClassificationSolutionImpactValuesCalculator.cs (modified) (2 diffs)

branches/HiveStatistics/sources

@@ -23 +23 @@
 bin
 protoc.exe
+obj

@@ -23 +23 @@
 bin
 protoc.exe
+obj

branches/HiveStatistics/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/Plugin.cs.frame

@@ -26 +26 @@
 
 namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Classification {
-  [Plugin("HeuristicLab.Problems.DataAnalysis.Symbolic.Classification","Provides classes to perform symbolic classification (single- or multiobjective).", "3.4.7.$WCREV$")]
+  [Plugin("HeuristicLab.Problems.DataAnalysis.Symbolic.Classification","Provides classes to perform symbolic classification (single- or multiobjective).", "3.4.8.$WCREV$")]
   [PluginFile("HeuristicLab.Problems.DataAnalysis.Symbolic.Classification-3.4.dll", PluginFileType.Assembly)]
   [PluginDependency("HeuristicLab.ALGLIB", "3.7.0")]

branches/HiveStatistics/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/Properties/AssemblyInfo.cs.frame

@@ -53 +53 @@
 // by using the '*' as shown below:
 [assembly: AssemblyVersion("3.4.0.0")]
-[assembly: AssemblyFileVersion("3.4.7.$WCREV$")]
+[assembly: AssemblyFileVersion("3.4.8.$WCREV$")]

branches/HiveStatistics/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/SymbolicClassificationPruningAnalyzer.cs

@@ -22 +22 @@
 using HeuristicLab.Common;
 using HeuristicLab.Core;
+using HeuristicLab.Data;
 using HeuristicLab.Parameters;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
@@ -45 +46 @@
 
     public SymbolicClassificationPruningAnalyzer() {
-      Parameters.Add(new ValueParameter<SymbolicDataAnalysisExpressionPruningOperator>(PruningOperatorParameterName, "The operator used to prune trees", new SymbolicClassificationPruningOperator(new SymbolicClassificationSolutionImpactValuesCalculator())));
+      Parameters.Add(new ValueParameter<SymbolicClassificationPruningOperator>(PruningOperatorParameterName, "The operator used to prune trees", new SymbolicClassificationPruningOperator(new SymbolicClassificationSolutionImpactValuesCalculator())));
+    }
+
+    [StorableHook(HookType.AfterDeserialization)]
+    private void AfterDeserialization() {
+      // BackwardsCompatibility3.3
+
+      #region Backwards compatible code, remove with 3.4
+      if (Parameters.ContainsKey(PruningOperatorParameterName)) {
+        var oldParam = Parameters[PruningOperatorParameterName] as ValueParameter<SymbolicDataAnalysisExpressionPruningOperator>;
+        if (oldParam != null) {
+          Parameters.Remove(oldParam);
+          Parameters.Add(new ValueParameter<SymbolicClassificationPruningOperator>(PruningOperatorParameterName, "The operator used to prune trees", new SymbolicClassificationPruningOperator(new SymbolicClassificationSolutionImpactValuesCalculator())));
+        }
+      } else {
+        // not yet contained
+        Parameters.Add(new ValueParameter<SymbolicClassificationPruningOperator>(PruningOperatorParameterName, "The operator used to prune trees", new SymbolicClassificationPruningOperator(new SymbolicClassificationSolutionImpactValuesCalculator())));
+      }
+
+      if (Parameters.ContainsKey("PruneOnlyZeroImpactNodes")) {
+        PruningOperator.PruneOnlyZeroImpactNodes = ((IFixedValueParameter<BoolValue>)Parameters["PruneOnlyZeroImpactNodes"]).Value.Value;
+        Parameters.Remove(Parameters["PruneOnlyZeroImpactNodes"]);
+      }
+      if (Parameters.ContainsKey("ImpactThreshold")) {
+        PruningOperator.NodeImpactThreshold = ((IFixedValueParameter<DoubleValue>)Parameters["ImpactThreshold"]).Value.Value;
+        Parameters.Remove(Parameters["ImpactThreshold"]);
+      }
+      if (Parameters.ContainsKey("ImpactValuesCalculator")) {
+        PruningOperator.ImpactValuesCalculator = ((ValueParameter<SymbolicDataAnalysisSolutionImpactValuesCalculator>)Parameters["ImpactValuesCalculator"]).Value;
+        Parameters.Remove(Parameters["ImpactValuesCalculator"]);
+      }
+      #endregion
     }
   }

branches/HiveStatistics/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/SymbolicClassificationPruningOperator.cs

@@ -35 +35 @@
   public class SymbolicClassificationPruningOperator : SymbolicDataAnalysisExpressionPruningOperator {
     private const string ModelCreatorParameterName = "ModelCreator";
+    private const string EvaluatorParameterName = "Evaluator";
 
     #region parameter properties
     public ILookupParameter<ISymbolicClassificationModelCreator> ModelCreatorParameter {
       get { return (ILookupParameter<ISymbolicClassificationModelCreator>)Parameters[ModelCreatorParameterName]; }
+    }
+
+    public ILookupParameter<ISymbolicClassificationSingleObjectiveEvaluator> EvaluatorParameter {
+      get {
+        return (ILookupParameter<ISymbolicClassificationSingleObjectiveEvaluator>)Parameters[EvaluatorParameterName];
+      }
     }
     #endregion
@@ -51 +58 @@
       : base(impactValuesCalculator) {
       Parameters.Add(new LookupParameter<ISymbolicClassificationModelCreator>(ModelCreatorParameterName));
+      Parameters.Add(new LookupParameter<ISymbolicClassificationSingleObjectiveEvaluator>(EvaluatorParameterName));
+    }
+
+    [StorableHook(HookType.AfterDeserialization)]
+    private void AfterDeserialization() {
+      // BackwardsCompatibility3.3
+      #region Backwards compatible code, remove with 3.4
+      base.ImpactValuesCalculator = new SymbolicClassificationSolutionImpactValuesCalculator();
+      if (!Parameters.ContainsKey(EvaluatorParameterName)) {
+        Parameters.Add(new LookupParameter<ISymbolicClassificationSingleObjectiveEvaluator>(EvaluatorParameterName));
+      }
+      #endregion
     }
 
@@ -62 +81 @@
 
     protected override double Evaluate(IDataAnalysisModel model) {
-      var classificationModel = (IClassificationModel)model;
+      var evaluator = EvaluatorParameter.ActualValue;
+      var classificationModel = (ISymbolicClassificationModel)model;
       var classificationProblemData = (IClassificationProblemData)ProblemDataParameter.ActualValue;
       var rows = Enumerable.Range(FitnessCalculationPartitionParameter.ActualValue.Start, FitnessCalculationPartitionParameter.ActualValue.Size);
-
-      return Evaluate(classificationModel, classificationProblemData, rows);
-    }
-
-    private static double Evaluate(IClassificationModel model, IClassificationProblemData problemData, IEnumerable<int> rows) {
-      var estimatedValues = model.GetEstimatedClassValues(problemData.Dataset, rows);
-      var targetValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, rows);
-      OnlineCalculatorError errorState;
-      var quality = OnlineAccuracyCalculator.Calculate(targetValues, estimatedValues, out errorState);
-      if (errorState != OnlineCalculatorError.None) return double.NaN;
-      return quality;
+      return evaluator.Evaluate(this.ExecutionContext, classificationModel.SymbolicExpressionTree, classificationProblemData, rows);
     }
 
@@ -86 +96 @@
 
       var nodes = clonedTree.Root.GetSubtree(0).GetSubtree(0).IterateNodesPrefix().ToList();
-      double quality = Evaluate(model, problemData, rows);
+      double qualityForImpactsCalculation = double.NaN;
 
       for (int i = 0; i < nodes.Count; ++i) {
@@ -92 +102 @@
         if (node is ConstantTreeNode) continue;
 
-        double impactValue, replacementValue;
-        impactValuesCalculator.CalculateImpactAndReplacementValues(model, node, problemData, rows, out impactValue, out replacementValue, quality);
+        double impactValue, replacementValue, newQualityForImpactsCalculation;
+        impactValuesCalculator.CalculateImpactAndReplacementValues(model, node, problemData, rows, out impactValue, out replacementValue, out newQualityForImpactsCalculation, qualityForImpactsCalculation);
 
         if (pruneOnlyZeroImpactNodes && !impactValue.IsAlmost(0.0)) continue;
@@ -104 +114 @@
         i += node.GetLength() - 1; // skip subtrees under the node that was folded
 
-        quality -= impactValue;
+        qualityForImpactsCalculation = newQualityForImpactsCalculation;
       }
       return model.SymbolicExpressionTree;

branches/HiveStatistics/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Classification/3.4/SymbolicClassificationSolutionImpactValuesCalculator.cs

@@ -47 +47 @@
     }
 
-    public override double CalculateImpactValue(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node, IDataAnalysisProblemData problemData, IEnumerable<int> rows, double originalQuality = double.NaN) {
+    public override double CalculateImpactValue(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node, IDataAnalysisProblemData problemData, IEnumerable<int> rows, double qualityForImpactsCalculation = double.NaN) {
       double impactValue, replacementValue;
-      CalculateImpactAndReplacementValues(model, node, problemData, rows, out impactValue, out replacementValue, originalQuality);
+      double newQualityForImpactsCalculation;
+      CalculateImpactAndReplacementValues(model, node, problemData, rows, out impactValue, out replacementValue, out newQualityForImpactsCalculation, qualityForImpactsCalculation);
       return impactValue;
     }
 
     public override void CalculateImpactAndReplacementValues(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node,
-      IDataAnalysisProblemData problemData, IEnumerable<int> rows, out double impactValue, out double replacementValue,
-      double originalQuality = Double.NaN) {
+      IDataAnalysisProblemData problemData, IEnumerable<int> rows, out double impactValue, out double replacementValue, out double newQualityForImpactsCalculation,
+      double qualityForImpactsCalculation = Double.NaN) {
       var classificationModel = (ISymbolicClassificationModel)model;
       var classificationProblemData = (IClassificationProblemData)problemData;
 
-      var dataset = classificationProblemData.Dataset;
-      var targetClassValues = dataset.GetDoubleValues(classificationProblemData.TargetVariable, rows);
-
-      OnlineCalculatorError errorState;
-      if (double.IsNaN(originalQuality)) {
-        var originalClassValues = classificationModel.GetEstimatedClassValues(dataset, rows);
-        originalQuality = OnlineAccuracyCalculator.Calculate(targetClassValues, originalClassValues, out errorState);
-        if (errorState != OnlineCalculatorError.None) originalQuality = 0.0;
-      }
+      if (double.IsNaN(qualityForImpactsCalculation))
+        qualityForImpactsCalculation = CalculateQualityForImpacts(classificationModel, classificationProblemData, rows);
 
       replacementValue = CalculateReplacementValue(classificationModel, node, classificationProblemData, rows);
@@ -81 +75 @@
       tempModelParentNode.InsertSubtree(i, constantNode);
 
+      OnlineCalculatorError errorState;
+      var dataset = classificationProblemData.Dataset;
+      var targetClassValues = dataset.GetDoubleValues(classificationProblemData.TargetVariable, rows);
       var estimatedClassValues = tempModel.GetEstimatedClassValues(dataset, rows);
-      double newQuality = OnlineAccuracyCalculator.Calculate(targetClassValues, estimatedClassValues, out errorState);
-      if (errorState != OnlineCalculatorError.None) newQuality = 0.0;
+      newQualityForImpactsCalculation = OnlineAccuracyCalculator.Calculate(targetClassValues, estimatedClassValues, out errorState);
+      if (errorState != OnlineCalculatorError.None) newQualityForImpactsCalculation = 0.0;
 
-      impactValue = originalQuality - newQuality;
+      impactValue = qualityForImpactsCalculation - newQualityForImpactsCalculation;
+    }
+
+    public static double CalculateQualityForImpacts(ISymbolicClassificationModel model, IClassificationProblemData problemData, IEnumerable<int> rows) {
+      OnlineCalculatorError errorState;
+      var dataset = problemData.Dataset;
+      var targetClassValues = dataset.GetDoubleValues(problemData.TargetVariable, rows);
+      var originalClassValues = model.GetEstimatedClassValues(dataset, rows);
+      var qualityForImpactsCalculation = OnlineAccuracyCalculator.Calculate(targetClassValues, originalClassValues, out errorState);
+      if (errorState != OnlineCalculatorError.None) qualityForImpactsCalculation = 0.0;
+
+      return qualityForImpactsCalculation;
     }
   }

@@ -23 +23 @@
 bin
 protoc.exe
+obj