Changeset 7672 for trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators

Timestamp:

03/30/12 10:21:21 (13 years ago)

Author:

mkommend

Message:

#1788: Prepared symbolic regression evaluators to apply linear scaling and moved to separate folder.

Location:

trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators

Files:

• . (added)
• SymbolicRegressionConstantOptimizationEvaluator.cs (moved) (moved from trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/SymbolicRegressionConstantOptimizationEvaluator.cs)
• SymbolicRegressionSingleObjectiveEvaluator.cs (moved) (moved from trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/SymbolicRegressionSingleObjectiveEvaluator.cs) (1 diff)
• SymbolicRegressionSingleObjectiveMeanSquaredErrorEvaluator.cs (moved) (moved from trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/SymbolicRegressionSingleObjectiveMeanSquaredErrorEvaluator.cs) (6 diffs)
• SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.cs (moved) (moved from trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.cs)

trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectiveEvaluator.cs

@@ -22 +22 @@
 
 using HeuristicLab.Common;
+using HeuristicLab.Core;
+using HeuristicLab.Data;
+using HeuristicLab.Parameters;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
 namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Regression {
+  [StorableClass]
   public abstract class SymbolicRegressionSingleObjectiveEvaluator : SymbolicDataAnalysisSingleObjectiveEvaluator<IRegressionProblemData>, ISymbolicRegressionSingleObjectiveEvaluator {
+    private const string ApplyLinearScalingParameterName = "ApplyLinearScaling";
+    public IFixedValueParameter<BoolValue> ApplyLinearScalingParameter {
+      get { return (IFixedValueParameter<BoolValue>)Parameters[ApplyLinearScalingParameterName]; }
+    }
+    public bool ApplyLinearScaling {
+      get { return ApplyLinearScalingParameter.Value.Value; }
+      set { ApplyLinearScalingParameter.Value.Value = value; }
+    }
+
     [StorableConstructor]
     protected SymbolicRegressionSingleObjectiveEvaluator(bool deserializing) : base(deserializing) { }
-    protected SymbolicRegressionSingleObjectiveEvaluator(SymbolicRegressionSingleObjectiveEvaluator original, Cloner cloner)
-      : base(original, cloner) {
+    protected SymbolicRegressionSingleObjectiveEvaluator(SymbolicRegressionSingleObjectiveEvaluator original, Cloner cloner) : base(original, cloner) { }
+    protected SymbolicRegressionSingleObjectiveEvaluator()
+      : base() {
+      Parameters.Add(new FixedValueParameter<BoolValue>(ApplyLinearScalingParameterName, "Flag that indicates if the individual should be linearly scaled before evaluating.", new BoolValue(true)));
+      ApplyLinearScalingParameter.Hidden = true;
     }
 
-    protected SymbolicRegressionSingleObjectiveEvaluator() : base() { }
+    [StorableHook(HookType.AfterDeserialization)]
+    private void AfterDeserialization() {
+      if (!Parameters.ContainsKey(ApplyLinearScalingParameterName)) {
+        Parameters.Add(new FixedValueParameter<BoolValue>(ApplyLinearScalingParameterName, "Flag that indicates if the individual should be linearly scaled before evaluating.", new BoolValue(false)));
+        ApplyLinearScalingParameter.Hidden = true;
+      }
+    }
   }
 }

trunk/sources/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectiveMeanSquaredErrorEvaluator.cs

@@ -20 +20 @@
 #endregion
 
+using System;
 using System.Collections.Generic;
+using System.Linq;
 using HeuristicLab.Common;
 using HeuristicLab.Core;
@@ -31 +33 @@
   [StorableClass]
   public class SymbolicRegressionSingleObjectiveMeanSquaredErrorEvaluator : SymbolicRegressionSingleObjectiveEvaluator {
+    public override bool Maximization { get { return false; } }
+    [ThreadStatic]
+    private static double[] estimatedValuesCache;
+
     [StorableConstructor]
     protected SymbolicRegressionSingleObjectiveMeanSquaredErrorEvaluator(bool deserializing) : base(deserializing) { }
@@ -39 +45 @@
       return new SymbolicRegressionSingleObjectiveMeanSquaredErrorEvaluator(this, cloner);
     }
-
     public SymbolicRegressionSingleObjectiveMeanSquaredErrorEvaluator() : base() { }
-
-    public override bool Maximization { get { return false; } }
 
     public override IOperation Apply() {
@@ -48 +51 @@
       IEnumerable<int> rows = GenerateRowsToEvaluate();
 
-      double quality = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, rows);
+      double quality = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, rows, ApplyLinearScaling);
       QualityParameter.ActualValue = new DoubleValue(quality);
 
@@ -54 +57 @@
     }
 
-    public static double Calculate(ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, ISymbolicExpressionTree solution, double lowerEstimationLimit, double upperEstimationLimit, IRegressionProblemData problemData, IEnumerable<int> rows) {
+    public static double Calculate(ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, ISymbolicExpressionTree solution, double lowerEstimationLimit, double upperEstimationLimit, IRegressionProblemData problemData, IEnumerable<int> rows, bool applyLinearScaling) {
       IEnumerable<double> estimatedValues = interpreter.GetSymbolicExpressionTreeValues(solution, problemData.Dataset, rows);
       IEnumerable<double> originalValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, rows);
-      IEnumerable<double> boundedEstimationValues = estimatedValues.LimitToRange(lowerEstimationLimit, upperEstimationLimit);
+      IEnumerable<double> boundedEstimatedValues = estimatedValues.LimitToRange(lowerEstimationLimit, upperEstimationLimit);
       OnlineCalculatorError errorState;
-      double mse = OnlineMeanSquaredErrorCalculator.Calculate(originalValues, boundedEstimationValues, out errorState);
-      if (errorState != OnlineCalculatorError.None) return double.NaN;
+
+      double mse;
+      if (applyLinearScaling) {
+        //int i = 0;
+        //int rowCount = rows.Count();
+        //if (estimatedValuesCache == null) estimatedValuesCache = new double[problemData.Dataset.Rows];
+        //foreach (var value in boundedEstimatedValues) {
+        //  estimatedValuesCache[i] = value;
+        //  i++;
+        //}
+
+        double alpha, beta;
+        OnlineLinearScalingParameterCalculator.Calculate(boundedEstimatedValues, originalValues, out alpha, out beta, out errorState);
+        mse = OnlineMeanSquaredErrorCalculator.Calculate(originalValues, boundedEstimatedValues.Select(value => value * beta + alpha), out errorState);
+      } else
+        mse = OnlineMeanSquaredErrorCalculator.Calculate(originalValues, boundedEstimatedValues, out errorState);
+
+      if (errorState != OnlineCalculatorError.None) return Double.NaN;
       else return mse;
     }
@@ -68 +87 @@
       EstimationLimitsParameter.ExecutionContext = context;
 
-      double mse = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows);
-
+      double mse = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows, ApplyLinearScaling);
 
       SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = null;