Changeset 18027 for branches/3026_IntegrationIntoSymSpace/HeuristicLab.Problems.DataAnalysis/3.4/Implementation

Timestamp:

07/20/21 18:13:55 (4 years ago)

Author:

dpiringe

Message:

#3026

• merged trunk into branch

Location:

branches/3026_IntegrationIntoSymSpace

Files:

• . (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis/3.4 (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/ClassificationSolutionBase.cs (modified) (2 diffs)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/DiscriminantFunctionClassificationSolutionBase.cs (modified) (6 diffs)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Interval/Interval.cs (modified) (4 diffs, 1 prop)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Interval/ShapeConstraint.cs (deleted)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Interval/ShapeConstraintsParser.cs (deleted)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionProblemData.cs (modified) (7 diffs)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/ShapeConstrainedRegressionProblem.cs (copied) (copied from trunk/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/ShapeConstrainedRegressionProblem.cs)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/ShapeConstrainedRegressionProblemData.cs (copied) (copied from trunk/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/ShapeConstrainedRegressionProblemData.cs)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/ShapeConstraint.cs (copied) (copied from trunk/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/ShapeConstraint.cs)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/ShapeConstraints.cs (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/ShapeConstraintsParser.cs (copied) (copied from trunk/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/ShapeConstraintsParser.cs)

branches/3026_IntegrationIntoSymSpace/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/ClassificationSolutionBase.cs

@@ -34 +34 @@
     private const string TrainingAccuracyResultName = "Accuracy (training)";
     private const string TestAccuracyResultName = "Accuracy (test)";
-    private const string TrainingNormalizedGiniCoefficientResultName = "Normalized Gini Coefficient (training)";
-    private const string TestNormalizedGiniCoefficientResultName = "Normalized Gini Coefficient (test)";
+    private const string TrainingNormalizedGiniCoefficientResultName = "Norm. Gini coeff. (training)";
+    private const string TestNormalizedGiniCoefficientResultName = "Norm. Gini coeff. (test)";
     private const string ClassificationPerformanceMeasuresResultName = "Classification Performance Measures";
 
@@ -97 +97 @@
       if (string.IsNullOrEmpty(Model.TargetVariable))
         Model.TargetVariable = this.ProblemData.TargetVariable;
-
-      if (!this.ContainsKey(TrainingNormalizedGiniCoefficientResultName))
+      var newResult = false;
+      if (!this.ContainsKey(TrainingNormalizedGiniCoefficientResultName)) {
         Add(new Result(TrainingNormalizedGiniCoefficientResultName, "Normalized Gini coefficient of the model on the training partition.", new DoubleValue()));
-      if (!this.ContainsKey(TestNormalizedGiniCoefficientResultName))
+        newResult = true;
+      }
+      if (!this.ContainsKey(TestNormalizedGiniCoefficientResultName)) {
         Add(new Result(TestNormalizedGiniCoefficientResultName, "Normalized Gini coefficient of the model on the test partition.", new DoubleValue()));
+        newResult = true;
+      }
       if (!this.ContainsKey(ClassificationPerformanceMeasuresResultName)) {
         Add(new Result(ClassificationPerformanceMeasuresResultName, @"Classification performance measures.\n
                               In a multiclass classification all misclassifications of the negative class will be treated as true negatives except on positive class estimations.",
                               new ClassificationPerformanceMeasuresResultCollection()));
-        CalculateClassificationResults();
+        newResult = true;
       }
+      if (newResult) CalculateClassificationResults();
     }
 

branches/3026_IntegrationIntoSymSpace/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/DiscriminantFunctionClassificationSolutionBase.cs

@@ -40 +40 @@
     private const string TrainingRSquaredResultName = "Pearson's R² (training)";
     private const string TestRSquaredResultName = "Pearson's R² (test)";
+    private const string TrainingNormalizedGiniCoefficientResultName = "Norm. Gini coeff. (training, discriminant values)";
+    private const string TestNormalizedGiniCoefficientResultName = "Norm. Gini coeff. (test, discriminant values)";
+
 
     public new IDiscriminantFunctionClassificationModel Model {
@@ -71 +74 @@
       private set { ((DoubleValue)this[TestRSquaredResultName].Value).Value = value; }
     }
+    public double TrainingNormalizedGiniCoefficientForDiscriminantValues {
+      get { return ((DoubleValue)this[TrainingNormalizedGiniCoefficientResultName].Value).Value; }
+      protected set { ((DoubleValue)this[TrainingNormalizedGiniCoefficientResultName].Value).Value = value; }
+    }
+    public double TestNormalizedGiniCoefficientForDiscriminantValues {
+      get { return ((DoubleValue)this[TestNormalizedGiniCoefficientResultName].Value).Value; }
+      protected set { ((DoubleValue)this[TestNormalizedGiniCoefficientResultName].Value).Value = value; }
+    }
     #endregion
 
@@ -85 +96 @@
       Add(new Result(TrainingRSquaredResultName, "Squared Pearson's correlation coefficient of the model output and the actual values on the training partition", new DoubleValue()));
       Add(new Result(TestRSquaredResultName, "Squared Pearson's correlation coefficient of the model output and the actual values on the test partition", new DoubleValue()));
+      Add(new Result(TrainingNormalizedGiniCoefficientResultName, "Normalized Gini coefficient of the discriminant values produced by the model on the training partition.", new DoubleValue()));
+      Add(new Result(TestNormalizedGiniCoefficientResultName, "Normalized Gini coefficient of the discriminant values produced by the model on the test partition.", new DoubleValue()));
       RegisterEventHandler();
     }
@@ -90 +103 @@
     [StorableHook(HookType.AfterDeserialization)]
     private void AfterDeserialization() {
+      #region backwards compatibility
+      if (!ContainsKey(TrainingNormalizedGiniCoefficientResultName)) {
+        Add(new Result(TrainingNormalizedGiniCoefficientResultName, "Normalized Gini coefficient of the discriminant values produced by the model on the training partition.", new DoubleValue()));
+        Add(new Result(TestNormalizedGiniCoefficientResultName, "Normalized Gini coefficient of the discriminant values produced by the model on the test partition.", new DoubleValue()));
+        double[] estimatedTrainingValues = EstimatedTrainingValues.ToArray(); // cache values
+        double[] originalTrainingValues = ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndices).ToArray();
+        double[] estimatedTestValues = EstimatedTestValues.ToArray(); // cache values
+        double[] originalTestValues = ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TestIndices).ToArray();
+        double trainingNormalizedGini = NormalizedGiniCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out var errorState);
+        if (errorState != OnlineCalculatorError.None) trainingNormalizedGini = double.NaN;
+        double testNormalizedGini = NormalizedGiniCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
+        if (errorState != OnlineCalculatorError.None) testNormalizedGini = double.NaN;
+
+        TrainingNormalizedGiniCoefficientForDiscriminantValues = trainingNormalizedGini;
+        TestNormalizedGiniCoefficientForDiscriminantValues = testNormalizedGini;
+      }
+      #endregion
       RegisterEventHandler();
     }
@@ -106 +136 @@
 
       double trainingR = OnlinePearsonsRCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
-      TrainingRSquared = errorState == OnlineCalculatorError.None ? trainingR*trainingR : double.NaN;
+      TrainingRSquared = errorState == OnlineCalculatorError.None ? trainingR * trainingR : double.NaN;
       double testR = OnlinePearsonsRCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
-      TestRSquared = errorState == OnlineCalculatorError.None ? testR*testR : double.NaN;
+      TestRSquared = errorState == OnlineCalculatorError.None ? testR * testR : double.NaN;
 
       double trainingNormalizedGini = NormalizedGiniCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
@@ -115 +145 @@
       if (errorState != OnlineCalculatorError.None) testNormalizedGini = double.NaN;
 
-      TrainingNormalizedGiniCoefficient = trainingNormalizedGini;
-      TestNormalizedGiniCoefficient = testNormalizedGini;
+      TrainingNormalizedGiniCoefficientForDiscriminantValues = trainingNormalizedGini;
+      TestNormalizedGiniCoefficientForDiscriminantValues = testNormalizedGini;
     }
 

branches/3026_IntegrationIntoSymSpace/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Interval/Interval.cs

@@ -78 +78 @@
       if (double.IsNegativeInfinity(LowerBound) && double.IsPositiveInfinity(UpperBound)) return true;
       if (other.LowerBound >= LowerBound && other.UpperBound <= UpperBound) return true;
-  
+
       return false;
     }
@@ -97 +97 @@
     /// </summary>
     public bool IsPositive {
-      get => LowerBound > 0.0; 
+      get => LowerBound > 0.0;
     }
 
@@ -130 +130 @@
         return false;
 
-      return (UpperBound==other.UpperBound || (double.IsNaN(UpperBound) && double.IsNaN(other.UpperBound)))
-        && (LowerBound==other.LowerBound || (double.IsNaN(LowerBound) && double.IsNaN(other.LowerBound)));
+      return (UpperBound == other.UpperBound || (double.IsNaN(UpperBound) && double.IsNaN(other.UpperBound)))
+        && (LowerBound == other.LowerBound || (double.IsNaN(LowerBound) && double.IsNaN(other.LowerBound)));
     }
 
@@ -238 +238 @@
     public static Interval Cube(Interval a) {
       return new Interval(Math.Pow(a.LowerBound, 3), Math.Pow(a.UpperBound, 3));
+    }
+
+    public static Interval Power(Interval a, int b) {
+      if (b < 0) return Power(1.0 / a, -b); // a^(-b) = 1/(a^b)
+      if (b == 0 && (a.Contains(0.0) || a.IsInfiniteOrUndefined)) return new Interval(double.NaN, double.NaN);  // 0^0, +/-inf^0 are undefined
+      if (b == 0) return new Interval(1.0, 1.0); // x^0 = 1
+      if (b == 1) return a;
+      if (b % 2 == 0) {
+        // even powers (see x²)
+        if (a.UpperBound <= 0) return new Interval(Math.Pow(a.UpperBound, b), Math.Pow(a.LowerBound, b));     // interval is negative
+        if (a.LowerBound >= 0) return new Interval(Math.Pow(a.LowerBound, b), Math.Pow(a.UpperBound, b)); // interval is positive
+        return new Interval(0, Math.Max(Math.Pow(a.LowerBound, b), Math.Pow(a.UpperBound, b))); // interval goes over zero
+      } else {
+        // odd powers (see x³)
+        return new Interval(Math.Pow(a.LowerBound, b), Math.Pow(a.UpperBound, b));
+      }
     }
 

branches/3026_IntegrationIntoSymSpace/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionProblemData.cs

@@ -35 +35 @@
     protected const string TargetVariableParameterName = "TargetVariable";
     protected const string VariableRangesParameterName = "VariableRanges";
-    protected const string ShapeConstraintsParameterName = "ShapeConstraints";
     public string Filename { get; set; }
 
@@ -78 +77 @@
       defaultDataset = new Dataset(new string[] { "y", "x" }, kozaF1);
       defaultDataset.Name = "Fourth-order Polynomial Function Benchmark Dataset";
-      defaultDataset.Description = "f(x) = x^4 + x^3 + x^2 + x^1";
+      defaultDataset.Description = "f(x) = x^4 + x^3 + x^2 + x";
       defaultAllowedInputVariables = new List<string>() { "x" };
       defaultTargetVariable = "y";
@@ -94 +93 @@
       problemData.Parameters.Add(new ConstrainedValueParameter<StringValue>(TargetVariableParameterName, new ItemSet<StringValue>()));
       problemData.Parameters.Add(new FixedValueParameter<IntervalCollection>(VariableRangesParameterName, "", new IntervalCollection()));
-      problemData.Parameters.Add(new FixedValueParameter<ShapeConstraints>(ShapeConstraintsParameterName, "", new ShapeConstraints()));
       emptyProblemData = problemData;
     }
@@ -101 +99 @@
     #region parameter properties
     public IConstrainedValueParameter<StringValue> TargetVariableParameter => (IConstrainedValueParameter<StringValue>)Parameters[TargetVariableParameterName];
-    public IFixedValueParameter<ShapeConstraints> ShapeConstraintsParameter => (IFixedValueParameter<ShapeConstraints>)Parameters[ShapeConstraintsParameterName];
     public IFixedValueParameter<IntervalCollection> VariableRangesParameter => (IFixedValueParameter<IntervalCollection>)Parameters[VariableRangesParameterName];
     #endregion
@@ -109 +106 @@
       get => VariableRangesParameter.Value;
     }
-
-
-    public ShapeConstraints ShapeConstraints => ShapeConstraintsParameter.Value;
-
 
     public string TargetVariable {
@@ -137 +130 @@
     private void AfterDeserialization() {
       if (!Parameters.ContainsKey(VariableRangesParameterName)) {
-        var intervalCollection = Dataset.GetIntervals();
-        Parameters.Add(new FixedValueParameter<IntervalCollection>(VariableRangesParameterName, intervalCollection));
-      }
-      if (Parameters.ContainsKey("IntervalConstraints")) {
-        var param = (IFixedValueParameter<ShapeConstraints>)Parameters["IntervalConstraints"];
-        Parameters.Remove(param);
-        Parameters.Add(new FixedValueParameter<ShapeConstraints>(ShapeConstraintsParameterName, param.Value));
-      }
-      if (!Parameters.ContainsKey(ShapeConstraintsParameterName)) {
-        Parameters.Add(new FixedValueParameter<ShapeConstraints>(ShapeConstraintsParameterName, new ShapeConstraints()));
+        var variableRanges = Dataset.GetVariableRanges();
+        Parameters.Add(new FixedValueParameter<IntervalCollection>(VariableRangesParameterName, variableRanges));
       }
 
@@ -174 +159 @@
     public RegressionProblemData(IDataset dataset, IEnumerable<string> allowedInputVariables, string targetVariable,
       IEnumerable<ITransformation> transformations = null,
-      IntervalCollection variableRanges = null,
-      ShapeConstraints shapeConstraints = null)
+      IntervalCollection variableRanges = null)
       : base(dataset, allowedInputVariables, transformations ?? Enumerable.Empty<ITransformation>()) {
       var variables = InputVariables.Select(x => x.AsReadOnly()).ToList();
       Parameters.Add(new ConstrainedValueParameter<StringValue>(TargetVariableParameterName, new ItemSet<StringValue>(variables), variables.Where(x => x.Value == targetVariable).First()));
       if (variableRanges == null) {
-        variableRanges = Dataset.GetIntervals();
+        variableRanges = Dataset.GetVariableRanges();
       }
       Parameters.Add(new FixedValueParameter<IntervalCollection>(VariableRangesParameterName, variableRanges));
-
-      if (shapeConstraints == null) {
-        shapeConstraints = new ShapeConstraints();
-      }
-      Parameters.Add(new FixedValueParameter<ShapeConstraints>(ShapeConstraintsParameterName, shapeConstraints));
-      RegisterParameterEvents();
     }
     private void RegisterParameterEvents() {
       TargetVariableParameter.ValueChanged += new EventHandler(Parameter_ValueChanged);
-      // VariableRanges and ShapeConstraints are fixed parameters
+      // VariableRanges are fixed parameters
     }
     private void Parameter_ValueChanged(object sender, EventArgs e) {

branches/3026_IntegrationIntoSymSpace/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/ShapeConstraints.cs

@@ -77 +77 @@
     public event EventHandler Changed;
 
+
     private void RaiseChanged() {
       var handlers = Changed;