Changeset 5271

Timestamp:

01/11/11 13:01:56 (14 years ago)

Author:

gkronber

Message:

Added training best solution analyzer for classification problems. #1369

Location:

trunk/sources/HeuristicLab.Problems.DataAnalysis.Classification/3.3

Files:

• HeuristicLab.Problems.DataAnalysis.Classification-3.3.csproj (modified) (2 diffs)
• Symbolic/Analyzer/TrainingBestSymbolicClassificationSolutionAnalyzer.cs (added)
• Symbolic/Analyzer/ValidationBestSymbolicClassificationSolutionAnalyzer.cs (modified) (9 diffs)
• Symbolic/SymbolicClassificationProblem.cs (modified) (2 diffs)

trunk/sources/HeuristicLab.Problems.DataAnalysis.Classification/3.3/HeuristicLab.Problems.DataAnalysis.Classification-3.3.csproj

@@ -12 +12 @@
     <AssemblyName>HeuristicLab.Problems.DataAnalysis.Classification-3.3</AssemblyName>
     <TargetFrameworkVersion>v4.0</TargetFrameworkVersion>
-    <TargetFrameworkProfile></TargetFrameworkProfile>
+    <TargetFrameworkProfile>
+    </TargetFrameworkProfile>
     <FileAlignment>512</FileAlignment>
     <TargetFrameworkProfile />
@@ -117 +118 @@
     <Compile Include="Properties\AssemblyInfo.cs" />
     <Compile Include="Interfaces\ISymbolicClassificationAnalyzer.cs" />
+    <Compile Include="Symbolic\Analyzer\TrainingBestSymbolicClassificationSolutionAnalyzer.cs" />
     <Compile Include="Symbolic\Analyzer\ValidationBestSymbolicClassificationSolutionAnalyzer.cs" />
     <Compile Include="Symbolic\Evaluators\OnlineAccuracyEvaluator.cs" />

trunk/sources/HeuristicLab.Problems.DataAnalysis.Classification/3.3/Symbolic/Analyzer/ValidationBestSymbolicClassificationSolutionAnalyzer.cs

@@ -52 +52 @@
     private const string UpperEstimationLimitParameterName = "UpperEstimationLimit";
     private const string LowerEstimationLimitParameterName = "LowerEstimationLimit";
+    private const string CalculateSolutionComplexityParameterName = "CalculateSolutionComplexity";
 
     private const string ResultsParameterName = "Results";
@@ -58 +59 @@
     private const string BestSolutionAccuracyTrainingParameterName = "Best solution accuracy (training)";
     private const string BestSolutionAccuracyTestParameterName = "Best solution accuracy (test)";
+    private const string BestSolutionLengthParameterName = "Best solution length (validation)";
+    private const string BestSolutionHeightParameterName = "Best solution height (validiation)";
     private const string VariableFrequenciesParameterName = "VariableFrequencies";
 
@@ -101 +104 @@
       get { return (ILookupParameter<DataTable>)Parameters[VariableFrequenciesParameterName]; }
     }
-
+    public IValueParameter<BoolValue> CalculateSolutionComplexityParameter {
+      get { return (IValueParameter<BoolValue>)Parameters[CalculateSolutionComplexityParameterName]; }
+    }
     public ILookupParameter<ResultCollection> ResultsParameter {
       get { return (ILookupParameter<ResultCollection>)Parameters[ResultsParameterName]; }
@@ -116 +121 @@
     public ILookupParameter<DoubleValue> BestSolutionAccuracyTestParameter {
       get { return (ILookupParameter<DoubleValue>)Parameters[BestSolutionAccuracyTestParameterName]; }
+    }
+    public ILookupParameter<IntValue> BestSolutionLengthParameter {
+      get { return (ILookupParameter<IntValue>)Parameters[BestSolutionLengthParameterName]; }
+    }
+    public ILookupParameter<IntValue> BestSolutionHeightParameter {
+      get { return (ILookupParameter<IntValue>)Parameters[BestSolutionHeightParameterName]; }
     }
     #endregion
@@ -159 +170 @@
       get { return VariableFrequenciesParameter.ActualValue; }
     }
+    public BoolValue CalculateSolutionComplexity {
+      get { return CalculateSolutionComplexityParameter.Value; }
+      set { CalculateSolutionComplexityParameter.Value = value; }
+    }
 
     public ResultCollection Results {
@@ -178 +193 @@
       get { return BestSolutionAccuracyTestParameter.ActualValue; }
       protected set { BestSolutionAccuracyTestParameter.ActualValue = value; }
+    }
+    public IntValue BestSolutionLength {
+      get { return BestSolutionLengthParameter.ActualValue; }
+      set { BestSolutionLengthParameter.ActualValue = value; }
+    }
+    public IntValue BestSolutionHeight {
+      get { return BestSolutionHeightParameter.ActualValue; }
+      set { BestSolutionHeightParameter.ActualValue = value; }
     }
     #endregion
@@ -202 +225 @@
       Parameters.Add(new ValueLookupParameter<DoubleValue>(LowerEstimationLimitParameterName, "The lower estimation limit that was set for the evaluation of the symbolic expression trees."));
       Parameters.Add(new LookupParameter<DataTable>(VariableFrequenciesParameterName, "The variable frequencies table to use for the calculation of variable impacts"));
-
+      Parameters.Add(new ValueParameter<BoolValue>(CalculateSolutionComplexityParameterName, "Determines if the length and height of the validation best solution should be calculated.", new BoolValue(true)));
       Parameters.Add(new ValueLookupParameter<ResultCollection>(ResultsParameterName, "The results collection where the analysis values should be stored."));
       Parameters.Add(new LookupParameter<DoubleValue>(BestValidationQualityParameterName, "The validation quality of the best solution in the current run."));
@@ -208 +231 @@
       Parameters.Add(new LookupParameter<DoubleValue>(BestSolutionAccuracyTrainingParameterName, "The training accuracy of the best solution."));
       Parameters.Add(new LookupParameter<DoubleValue>(BestSolutionAccuracyTestParameterName, "The test accuracy of the best solution."));
+      Parameters.Add(new LookupParameter<IntValue>(BestSolutionLengthParameterName, "The length of the best symbolic classification solution."));
+      Parameters.Add(new LookupParameter<IntValue>(BestSolutionHeightParameterName, "The height of the best symbolic classification solution."));
+    }
+
+    [StorableHook(HookType.AfterDeserialization)]
+    private void AfterDeserialization() {
+      if (!Parameters.ContainsKey(CalculateSolutionComplexityParameterName)) {
+        Parameters.Add(new ValueParameter<BoolValue>(CalculateSolutionComplexityParameterName, "Determines if the length and height of the validation best solution should be calculated.", new BoolValue(true)));
+      }
+      if (!Parameters.ContainsKey(BestSolutionLengthParameterName)) {
+        Parameters.Add(new LookupParameter<IntValue>(BestSolutionLengthParameterName, "The length of the best symbolic classification solution."));
+      }
+      if (!Parameters.ContainsKey(BestSolutionHeightParameterName)) {
+        Parameters.Add(new LookupParameter<IntValue>(BestSolutionHeightParameterName, "The height of the best symbolic classification solution."));
+      }
     }
 
@@ -278 +316 @@
 
     private void UpdateBestSolutionResults() {
+      if (CalculateSolutionComplexity.Value) {
+        BestSolutionLength = new IntValue(BestValidationSolution.Model.SymbolicExpressionTree.Size);
+        BestSolutionHeight = new IntValue(BestValidationSolution.Model.SymbolicExpressionTree.Height);
+        if (!Results.ContainsKey(BestSolutionLengthParameterName)) {
+          Results.Add(new Result(BestSolutionLengthParameterName, "Length of the best solution on the validation set", new IntValue()));
+          Results.Add(new Result(BestSolutionHeightParameterName, "Height of the best solution on the validation set", new IntValue()));
+        }
+        Results[BestSolutionLengthParameterName].Value = BestSolutionLength;
+        Results[BestSolutionHeightParameterName].Value = BestSolutionHeight;
+      }
+
       BestSymbolicRegressionSolutionAnalyzer.UpdateBestSolutionResults(BestValidationSolution, ClassificationProblemData, Results, Generations, VariableFrequencies);
 

trunk/sources/HeuristicLab.Problems.DataAnalysis.Classification/3.3/Symbolic/SymbolicClassificationProblem.cs

@@ -237 +237 @@
       Operators.Add(new SymbolicRegressionVariableFrequencyAnalyzer());
       Operators.Add(new ValidationBestSymbolicClassificationSolutionAnalyzer());
+      Operators.Add(new TrainingBestSymbolicClassificationSolutionAnalyzer());
     }
 
@@ -314 +315 @@
           bestValidationSolutionAnalyzer.ValidationSamplesEndParameter.Value = ValidationSamplesEnd;
         }
+        var bestTrainingSolutionAnalyzer = analyzer as TrainingBestSymbolicClassificationSolutionAnalyzer;
+        if (bestTrainingSolutionAnalyzer != null) {
+          bestTrainingSolutionAnalyzer.ProblemDataParameter.ActualName = ClassificationProblemDataParameter.Name;
+          bestTrainingSolutionAnalyzer.UpperEstimationLimitParameter.ActualName = UpperEstimationLimitParameter.Name;
+          bestTrainingSolutionAnalyzer.LowerEstimationLimitParameter.ActualName = LowerEstimationLimitParameter.Name;
+          bestTrainingSolutionAnalyzer.SymbolicExpressionTreeInterpreterParameter.ActualName = SymbolicExpressionTreeInterpreterParameter.Name;
+          bestTrainingSolutionAnalyzer.SymbolicExpressionTreeParameter.ActualName = SolutionCreator.SymbolicExpressionTreeParameter.ActualName;
+        }
         var varFreqAnalyzer = analyzer as SymbolicRegressionVariableFrequencyAnalyzer;
         if (varFreqAnalyzer != null) {