Changeset 12247 for branches/HeuristicLab.DatasetRefactor/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression

Timestamp:

03/24/15 11:17:08 (10 years ago)

Author:

mkommend

Message:

#2276: Merged trunk changes into dataset refactoring branch.

Location:

branches/HeuristicLab.DatasetRefactor/sources

Files:

• . (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Regression (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SymbolicRegressionPruningAnalyzer.cs (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SymbolicRegressionPruningOperator.cs (modified) (5 diffs)

branches/HeuristicLab.DatasetRefactor/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SymbolicRegressionPruningAnalyzer.cs

@@ -45 +45 @@
     public SymbolicRegressionPruningAnalyzer() {
       Parameters.Add(new ValueParameter<SymbolicDataAnalysisSolutionImpactValuesCalculator>(ImpactValuesCalculatorParameterName, "The impact values calculator", new SymbolicRegressionSolutionImpactValuesCalculator()));
-      Parameters.Add(new ValueParameter<SymbolicDataAnalysisExpressionPruningOperator>(PruningOperatorParameterName, "The operator used to prune trees", new SymbolicRegressionPruningOperator()));
+      Parameters.Add(new ValueParameter<SymbolicDataAnalysisExpressionPruningOperator>(PruningOperatorParameterName, "The operator used to prune trees", new SymbolicRegressionPruningOperator(new SymbolicRegressionSolutionImpactValuesCalculator())));
     }
   }

branches/HeuristicLab.DatasetRefactor/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SymbolicRegressionPruningOperator.cs

@@ -22 +22 @@
 #endregion
 
+using System.Collections.Generic;
 using System.Linq;
 using HeuristicLab.Common;
 using HeuristicLab.Core;
-using HeuristicLab.Parameters;
+using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
 
@@ -32 +33 @@
   [Item("SymbolicRegressionPruningOperator", "An operator which prunes symbolic regression trees.")]
   public class SymbolicRegressionPruningOperator : SymbolicDataAnalysisExpressionPruningOperator {
-    private const string ImpactValuesCalculatorParameterName = "ImpactValuesCalculator";
-
     protected SymbolicRegressionPruningOperator(SymbolicRegressionPruningOperator original, Cloner cloner)
       : base(original, cloner) {
@@ -44 +43 @@
     protected SymbolicRegressionPruningOperator(bool deserializing) : base(deserializing) { }
 
-    public SymbolicRegressionPruningOperator() {
-      var impactValuesCalculator = new SymbolicRegressionSolutionImpactValuesCalculator();
-      Parameters.Add(new ValueParameter<ISymbolicDataAnalysisSolutionImpactValuesCalculator>(ImpactValuesCalculatorParameterName, "The impact values calculator to be used for figuring out the node impacts.", impactValuesCalculator));
+    public SymbolicRegressionPruningOperator(ISymbolicDataAnalysisSolutionImpactValuesCalculator impactValuesCalculator)
+      : base(impactValuesCalculator) {
     }
 
-    protected override ISymbolicDataAnalysisModel CreateModel() {
-      return new SymbolicRegressionModel(SymbolicExpressionTree, Interpreter, EstimationLimits.Lower, EstimationLimits.Upper);
+    protected override ISymbolicDataAnalysisModel CreateModel(ISymbolicExpressionTree tree, ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, IDataAnalysisProblemData problemData, DoubleLimit estimationLimits) {
+      return new SymbolicRegressionModel(tree, interpreter, estimationLimits.Lower, estimationLimits.Upper);
     }
 
@@ -56 +54 @@
       var regressionModel = (IRegressionModel)model;
       var regressionProblemData = (IRegressionProblemData)ProblemData;
-      var trainingIndices = Enumerable.Range(FitnessCalculationPartition.Start, FitnessCalculationPartition.Size);
-      var estimatedValues = regressionModel.GetEstimatedValues(ProblemData.Dataset, trainingIndices); // also bounds the values
-      var targetValues = ProblemData.Dataset.GetDoubleValues(regressionProblemData.TargetVariable, trainingIndices);
+      var rows = Enumerable.Range(FitnessCalculationPartition.Start, FitnessCalculationPartition.Size);
+      return Evaluate(regressionModel, regressionProblemData, rows);
+    }
+
+    private static double Evaluate(IRegressionModel model, IRegressionProblemData problemData,
+      IEnumerable<int> rows) {
+      var estimatedValues = model.GetEstimatedValues(problemData.Dataset, rows); // also bounds the values
+      var targetValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, rows);
       OnlineCalculatorError errorState;
       var quality = OnlinePearsonsRSquaredCalculator.Calculate(targetValues, estimatedValues, out errorState);
@@ -64 +67 @@
       return quality;
     }
+
+    public static ISymbolicExpressionTree Prune(ISymbolicExpressionTree tree, SymbolicRegressionSolutionImpactValuesCalculator impactValuesCalculator, ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, IRegressionProblemData problemData, DoubleLimit estimationLimits, IEnumerable<int> rows, double nodeImpactThreshold = 0.0, bool pruneOnlyZeroImpactNodes = false) {
+      var clonedTree = (ISymbolicExpressionTree)tree.Clone();
+      var model = new SymbolicRegressionModel(clonedTree, interpreter, estimationLimits.Lower, estimationLimits.Upper);
+      var nodes = clonedTree.IterateNodesPrefix().ToList();
+      double quality = Evaluate(model, problemData, rows);
+
+      for (int i = 0; i < nodes.Count; ++i) {
+        var node = nodes[i];
+        if (node is ConstantTreeNode) continue;
+
+        double impactValue, replacementValue;
+        impactValuesCalculator.CalculateImpactAndReplacementValues(model, node, problemData, rows, out impactValue, out replacementValue, quality);
+
+        if (pruneOnlyZeroImpactNodes) {
+          if (!impactValue.IsAlmost(0.0)) continue;
+        } else if (nodeImpactThreshold < impactValue) {
+          continue;
+        }
+
+        var constantNode = (ConstantTreeNode)node.Grammar.GetSymbol("Constant").CreateTreeNode();
+        constantNode.Value = replacementValue;
+
+        ReplaceWithConstant(node, constantNode);
+        i += node.GetLength() - 1; // skip subtrees under the node that was folded
+
+        quality -= impactValue;
+      }
+      return model.SymbolicExpressionTree;
+    }
   }
 }