Changeset 4341 for branches/DataAnalysis/HeuristicLab.Problems.DataAnalysis.Regression

Timestamp:

08/30/10 10:04:03 (14 years ago)

Author:

gkronber

Message:

Merged changesets from revisions r4249, r4250, r4251, r4291, r4295 from trunk into data analysis exploration #1142.

Location:

branches/DataAnalysis

Files:

• . (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis.Regression/3.3/HeuristicLab.Problems.DataAnalysis.Regression-3.3.csproj (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis.Regression/3.3/Symbolic/SymbolicRegressionProblem.cs (modified) (2 diffs)
• HeuristicLab.Problems.DataAnalysis.Regression/3.3/Symbolic/SymbolicRegressionProblemBase.cs (modified) (5 diffs)
• HeuristicLab.Problems.DataAnalysis.Regression/3.3/Symbolic/SymbolicRegressionSolution.cs (modified) (2 diffs)

branches/DataAnalysis/HeuristicLab.Problems.DataAnalysis.Regression/3.3/HeuristicLab.Problems.DataAnalysis.Regression-3.3.csproj

@@ -127 +127 @@
     </Compile>
     <Compile Include="Symbolic\Analyzers\SymbolicRegressionSolutionLinearScaler.cs" />
-    <Compile Include="Symbolic\Analyzers\SymbolicRegressionTournamentPruning.cs" />
+    <Compile Include="Symbolic\Analyzers\SymbolicRegressionTournamentPruning.cs">
+      <SubType>Code</SubType>
+    </Compile>
     <Compile Include="Symbolic\Analyzers\SymbolicRegressionVariableFrequencyAnalyzer.cs" />
     <Compile Include="Symbolic\Analyzers\ValidationBestScaledSymbolicRegressionSolutionAnalyzer.cs" />

branches/DataAnalysis/HeuristicLab.Problems.DataAnalysis.Regression/3.3/Symbolic/SymbolicRegressionProblem.cs

@@ -21 +21 @@
 
 using System;
-using System.Collections.Generic;
 using System.Linq;
 using HeuristicLab.Common;
 using HeuristicLab.Core;
 using HeuristicLab.Data;
-using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
-using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding.Analyzers;
-using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding.Creators;
-using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding.Interfaces;
 using HeuristicLab.Optimization;
 using HeuristicLab.Parameters;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
-using HeuristicLab.PluginInfrastructure;
 using HeuristicLab.Problems.DataAnalysis.Regression.Symbolic.Analyzers;
-using HeuristicLab.Problems.DataAnalysis.Symbolic;
 
 namespace HeuristicLab.Problems.DataAnalysis.Regression.Symbolic {
@@ -84 +77 @@
     public SymbolicRegressionProblem()
       : base() {
-      var evaluator = new SymbolicRegressionScaledMeanSquaredErrorEvaluator();
-      Parameters.Add(new ValueParameter<BoolValue>("Maximization", "Set to false as the error of the regression model should be minimized.", (BoolValue)new BoolValue(false).AsReadOnly()));
+      var evaluator = new SymbolicRegressionPearsonsRSquaredEvaluator();
+      Parameters.Add(new ValueParameter<BoolValue>("Maximization", "Set to false as the error of the regression model should be minimized.", (BoolValue)new BoolValue(true)));
       Parameters.Add(new ValueParameter<ISymbolicRegressionEvaluator>("Evaluator", "The operator which should be used to evaluate symbolic regression solutions.", evaluator));
       Parameters.Add(new OptionalValueParameter<DoubleValue>("BestKnownQuality", "The minimal error value that reached by symbolic regression solutions for the problem."));
 
-      evaluator.QualityParameter.ActualName = "TrainingMeanSquaredError";
+      evaluator.QualityParameter.ActualName = "TrainingPearsonR2";
 
       InitializeOperators();

branches/DataAnalysis/HeuristicLab.Problems.DataAnalysis.Regression/3.3/Symbolic/SymbolicRegressionProblemBase.cs

@@ -113 +113 @@
     public ISymbolicExpressionGrammar FunctionTreeGrammar {
       get { return (ISymbolicExpressionGrammar)FunctionTreeGrammarParameter.Value; }
+      private set { FunctionTreeGrammarParameter.Value = value; }
     }
     public override IEnumerable<IOperator> Operators {
@@ -190 +191 @@
       MaxFunctionDefiningBranchesParameter.ValueChanged += new EventHandler(ArchitectureParameter_ValueChanged);
       SolutionCreatorParameter.ValueChanged += new EventHandler(SolutionCreatorParameter_ValueChanged);
+      FunctionTreeGrammarParameter.ValueChanged += new EventHandler(FunctionTreeGrammarParameter_ValueChanged);
     }
 
@@ -211 +213 @@
       UpdateGrammar();
     }
-    protected virtual void OnGrammarChanged(EventArgs e) { }
+    protected virtual void OnGrammarChanged() { UpdateGrammar(); }
     protected virtual void OnOperatorsChanged(EventArgs e) { RaiseOperatorsChanged(e); }
     protected virtual void OnSolutionCreatorChanged(EventArgs e) {
@@ -231 +233 @@
 
     #region event handlers
+    private void FunctionTreeGrammarParameter_ValueChanged(object sender, EventArgs e) {
+      if (!(FunctionTreeGrammar is GlobalSymbolicExpressionGrammar))
+        FunctionTreeGrammar = new GlobalSymbolicExpressionGrammar(FunctionTreeGrammar);
+      OnGrammarChanged();
+    }
+
     private void SolutionCreatorParameter_ValueChanged(object sender, EventArgs e) {
       OnSolutionCreatorChanged(e);
@@ -287 +295 @@
       operators = new List<IOperator>();
       operators.AddRange(ApplicationManager.Manager.GetInstances<ISymbolicExpressionTreeOperator>().OfType<IOperator>());
-      operators.Add(new SymbolicRegressionTournamentPruning());
       operators.Add(new SymbolicRegressionVariableFrequencyAnalyzer());
       operators.Add(new MinAverageMaxSymbolicExpressionTreeSizeAnalyzer());

branches/DataAnalysis/HeuristicLab.Problems.DataAnalysis.Regression/3.3/Symbolic/SymbolicRegressionSolution.cs

@@ -26 +26 @@
 using HeuristicLab.Core;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
+using HeuristicLab.Problems.DataAnalysis.Symbolic.Symbols;
 
 namespace HeuristicLab.Problems.DataAnalysis.Regression.Symbolic {
@@ -50 +51 @@
 
     protected override void RecalculateEstimatedValues() {
-      estimatedValues = (from x in Model.GetEstimatedValues(ProblemData, 0, ProblemData.Dataset.Rows)
-                         let boundedX = Math.Min(UpperEstimationLimit, Math.Max(LowerEstimationLimit, x))
-                         select double.IsNaN(boundedX) ? UpperEstimationLimit : boundedX).ToList();
+      int minLag = 0;
+      var laggedTreeNodes = Model.SymbolicExpressionTree.IterateNodesPrefix().OfType<LaggedVariableTreeNode>();
+      if (laggedTreeNodes.Any())
+        minLag = laggedTreeNodes.Min(node => node.Lag);
+      IEnumerable<double> calculatedValues =
+          from x in Model.GetEstimatedValues(ProblemData, 0 - minLag, ProblemData.Dataset.Rows)
+          let boundedX = Math.Min(UpperEstimationLimit, Math.Max(LowerEstimationLimit, x))
+          select double.IsNaN(boundedX) ? UpperEstimationLimit : boundedX;
+      estimatedValues = Enumerable.Repeat(double.NaN, Math.Abs(minLag)).Concat(calculatedValues).ToList();
       OnEstimatedValuesChanged();
     }