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

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 (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/Interfaces/ISymbolicDataAnalysisImpactValuesCalculator.cs (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/Plugin.cs.frame (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/Properties/AssemblyInfo.cs.frame (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/SymbolicDataAnalysisExpressionPruningOperator.cs (modified) (7 diffs)
• HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/SymbolicDataAnalysisSolutionImpactValuesCalculator.cs (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/TreeMatching/SymbolicExpressionTreePhenotypicSimilarityCalculator.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/3.4/Interfaces/ISymbolicDataAnalysisImpactValuesCalculator.cs

@@ -6 +6 @@
   public interface ISymbolicDataAnalysisSolutionImpactValuesCalculator : IItem {
     double CalculateReplacementValue(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node, IDataAnalysisProblemData problemData, IEnumerable<int> rows);
-    double CalculateImpactValue(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node, IDataAnalysisProblemData problemData, IEnumerable<int> rows, double originalQuality = double.NaN);
+    double CalculateImpactValue(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node, IDataAnalysisProblemData problemData, IEnumerable<int> rows, double qualityForImpactsCalculation = double.NaN);
     void CalculateImpactAndReplacementValues(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node, IDataAnalysisProblemData problemData,
-      IEnumerable<int> rows, out double impactValue, out double replacementValue, double originalQuality = double.NaN);
+      IEnumerable<int> rows, out double impactValue, out double replacementValue, out double newQualityForImpactsCalculation, double qualityForImpactsCalculation = double.NaN);
   }
 }

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

@@ -26 +26 @@
 
 namespace HeuristicLab.Problems.DataAnalysis.Symbolic {
-  [Plugin("HeuristicLab.Problems.DataAnalysis.Symbolic","Provides base classes for symbolic data analysis tasks.", "3.4.7.$WCREV$")]
+  [Plugin("HeuristicLab.Problems.DataAnalysis.Symbolic","Provides base classes for symbolic data analysis tasks.", "3.4.8.$WCREV$")]
   [PluginFile("HeuristicLab.Problems.DataAnalysis.Symbolic-3.4.dll", PluginFileType.Assembly)]
   [PluginDependency("HeuristicLab.ALGLIB", "3.7.0")]

branches/HiveStatistics/sources/HeuristicLab.Problems.DataAnalysis.Symbolic/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/3.4/SymbolicDataAnalysisExpressionPruningOperator.cs

@@ -49 +49 @@
     private const string EstimationLimitsParameterName = "EstimationLimits";
     private const string InterpreterParameterName = "SymbolicExpressionTreeInterpreter";
+    private const string ApplyLinearScalingParameterName = "ApplyLinearScaling";
     #endregion
 
@@ -87 +88 @@
     public ILookupParameter<ISymbolicDataAnalysisExpressionTreeInterpreter> InterpreterParameter {
       get { return (ILookupParameter<ISymbolicDataAnalysisExpressionTreeInterpreter>)Parameters[InterpreterParameterName]; }
+    }
+    public ILookupParameter<BoolValue> ApplyLinearScalingParameter {
+      get { return (ILookupParameter<BoolValue>)Parameters[ApplyLinearScalingParameterName]; }
     }
     #endregion
@@ -124 +128 @@
       Parameters.Add(new LookupParameter<ISymbolicExpressionTree>(SymbolicExpressionTreeParameterName));
       Parameters.Add(new LookupParameter<DoubleValue>(QualityParameterName));
+      Parameters.Add(new LookupParameter<BoolValue>(ApplyLinearScalingParameterName));
       Parameters.Add(new ValueParameter<ISymbolicDataAnalysisSolutionImpactValuesCalculator>(ImpactValuesCalculatorParameterName, impactValuesCalculator));
+      #endregion
+    }
+
+    [StorableHook(HookType.AfterDeserialization)]
+    private void AfterDeserialization() {
+      // BackwardsCompatibility3.3
+      #region Backwards compatible code, remove with 3.4
+      if (!Parameters.ContainsKey(PrunedNodesParameterName)) {
+        Parameters.Add(new LookupParameter<IntValue>(PrunedNodesParameterName, "A counter of how many nodes were pruned."));
+      }
+      if (!Parameters.ContainsKey(ApplyLinearScalingParameterName)) {
+        Parameters.Add(new LookupParameter<BoolValue>(ApplyLinearScalingParameterName));
+      }
+      if (!Parameters.ContainsKey(ImpactValuesCalculatorParameterName)) {
+        // value must be set by derived operators (regression/classification)
+        Parameters.Add(new ValueParameter<ISymbolicDataAnalysisSolutionImpactValuesCalculator>(ImpactValuesCalculatorParameterName));
+      }
       #endregion
     }
@@ -141 +163 @@
       var model = CreateModel(tree, interpreter, problemData, estimationLimits);
       var nodes = tree.Root.GetSubtree(0).GetSubtree(0).IterateNodesPrefix().ToList();
-      var rows = Enumerable.Range(fitnessCalculationPartition.Start, fitnessCalculationPartition.Size);
+      var rows = Enumerable.Range(fitnessCalculationPartition.Start, fitnessCalculationPartition.Size).ToList();
       var prunedSubtrees = 0;
       var prunedTrees = 0;
       var prunedNodes = 0;
 
-      double quality = Evaluate(model);
+      double qualityForImpactsCalculation = double.NaN;
 
       for (int i = 0; i < nodes.Count; ++i) {
@@ -153 +175 @@
 
         double impactValue, replacementValue;
-        ImpactValuesCalculator.CalculateImpactAndReplacementValues(model, node, problemData, rows, out impactValue, out replacementValue, quality);
+        double newQualityForImpacts;
+        ImpactValuesCalculator.CalculateImpactAndReplacementValues(model, node, problemData, rows, out impactValue, out replacementValue, out newQualityForImpacts, qualityForImpactsCalculation);
 
         if (PruneOnlyZeroImpactNodes && !impactValue.IsAlmost(0.0)) continue;
@@ -165 +188 @@
         i += length - 1; // skip subtrees under the node that was folded
 
-        quality -= impactValue;
         prunedSubtrees++;
         prunedNodes += length;
+
+        qualityForImpactsCalculation = newQualityForImpacts;
       }
 
@@ -174 +198 @@
       PrunedTreesParameter.ActualValue = new IntValue(prunedTrees);
       PrunedNodesParameter.ActualValue = new IntValue(prunedNodes);
-      QualityParameter.ActualValue.Value = quality;
+
+      if (prunedSubtrees > 0) // if nothing was pruned then there's no need to re-evaluate the tree
+        QualityParameter.ActualValue.Value = Evaluate(model);
 
       return base.Apply();

branches/HiveStatistics/sources/HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/SymbolicDataAnalysisSolutionImpactValuesCalculator.cs

@@ -37 +37 @@
     protected SymbolicDataAnalysisSolutionImpactValuesCalculator(bool deserializing) : base(deserializing) { }
     public abstract double CalculateReplacementValue(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node, IDataAnalysisProblemData problemData, IEnumerable<int> rows);
-    public abstract double CalculateImpactValue(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node, IDataAnalysisProblemData problemData, IEnumerable<int> rows, double originalQuality = double.NaN);
-    public abstract void CalculateImpactAndReplacementValues(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node, IDataAnalysisProblemData problemData, IEnumerable<int> rows, out double impactValue, out double replacementValue, double originalQuality = double.NaN);
+    public abstract double CalculateImpactValue(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node, IDataAnalysisProblemData problemData, IEnumerable<int> rows, double qualityForImpactsCalculation = double.NaN);
+    public abstract void CalculateImpactAndReplacementValues(ISymbolicDataAnalysisModel model, ISymbolicExpressionTreeNode node, IDataAnalysisProblemData problemData, IEnumerable<int> rows, out double impactValue, out double replacementValue, out double newQualityForImpactsCalculation, double qualityForImpactsCalculation = double.NaN);
 
     protected static double CalculateReplacementValue(ISymbolicExpressionTreeNode node, ISymbolicExpressionTree sourceTree, ISymbolicDataAnalysisExpressionTreeInterpreter interpreter,

branches/HiveStatistics/sources/HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/TreeMatching/SymbolicExpressionTreePhenotypicSimilarityCalculator.cs

@@ -67 +67 @@
       var r = OnlinePearsonsRCalculator.Calculate(v1, v2, out error);
 
-      if (r > 1.0)
-        r = 1.0;
+      var r2 = error == OnlineCalculatorError.None ? r * r : 0;
 
-      return error == OnlineCalculatorError.None ? r*r : 0;
+      if (r2 > 1.0)
+        r2 = 1.0;
+
+      return r2;
     }
 
@@ -89 +91 @@
       var r = OnlinePearsonsRCalculator.Calculate(leftValues, rightValues, out error);
 
-      if (r > 1.0)
-        r = 1.0; // sometimes due to fp errors it can happen that the correlation is over 1 (like 1.0000000009)
+      var r2 = error == OnlineCalculatorError.None ? r * r : 0;
 
-      return error == OnlineCalculatorError.None ? r*r : 0;
+      if (r2 > 1.0)
+        r2 = 1.0;
+
+      return r2;
     }
   }

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