Changeset 13948 for branches/HeuristicLab.RegressionSolutionGradientView/HeuristicLab.Algorithms.DataAnalysis/3.4/GaussianProcess

Timestamp:

06/29/16 10:36:52 (9 years ago)

Author:

pfleck

Message:

#2597

• Merged recent trunk changes.
• Adapted VariablesUsedForPrediction property for RegressionSolutionTargetResponseGradientView.
• Fixed a reference (.dll to project ref).

Location:

branches/HeuristicLab.RegressionSolutionGradientView/HeuristicLab.Algorithms.DataAnalysis

Files:

• . (modified) (1 prop)
• 3.4/GaussianProcess/CovarianceFunctions/CovarianceProduct.cs (modified) (1 diff)
• 3.4/GaussianProcess/CovarianceFunctions/CovarianceSum.cs (modified) (1 diff)
• 3.4/GaussianProcess/GaussianProcessModel.cs (modified) (7 diffs)
• 3.4/GaussianProcess/StudentTProcessModel.cs (modified) (6 diffs)

branches/HeuristicLab.RegressionSolutionGradientView/HeuristicLab.Algorithms.DataAnalysis/3.4/GaussianProcess/CovarianceFunctions/CovarianceProduct.cs

@@ -94 +94 @@
     public static IList<double> GetGradient(double[,] x, int i, int j, List<ParameterizedCovarianceFunction> factorFunctions) {
       var covariances = factorFunctions.Select(f => f.Covariance(x, i, j)).ToArray();
-      var gr = new List<double>(factorFunctions.Sum(f => f.CovarianceGradient(x, i, j).Count));
+      var gr = new List<double>();
       for (int ii = 0; ii < factorFunctions.Count; ii++) {
         foreach (var g in factorFunctions[ii].CovarianceGradient(x, i, j)) {

branches/HeuristicLab.RegressionSolutionGradientView/HeuristicLab.Algorithms.DataAnalysis/3.4/GaussianProcess/CovarianceFunctions/CovarianceSum.cs

@@ -88 +88 @@
       sum.CrossCovariance = (x, xt, i, j) => functions.Select(e => e.CrossCovariance(x, xt, i, j)).Sum();
       sum.CovarianceGradient = (x, i, j) => {
-        var g = new List<double>(functions.Sum(e => e.CovarianceGradient(x, i, j).Count));
+        var g = new List<double>();
         foreach (var e in functions)
           g.AddRange(e.CovarianceGradient(x, i, j));

branches/HeuristicLab.RegressionSolutionGradientView/HeuristicLab.Algorithms.DataAnalysis/3.4/GaussianProcess/GaussianProcessModel.cs

@@ -34 +34 @@
   [StorableClass]
   [Item("GaussianProcessModel", "Represents a Gaussian process posterior.")]
-  public sealed class GaussianProcessModel : NamedItem, IGaussianProcessModel {
+  public sealed class GaussianProcessModel : RegressionModel, IGaussianProcessModel {
+    public override IEnumerable<string> VariablesUsedForPrediction {
+      get { return allowedInputVariables; }
+    }
+
     [Storable]
     private double negativeLogLikelihood;
@@ -61 +65 @@
       get { return meanFunction; }
     }
-    [Storable]
-    private string targetVariable;
-    public string TargetVariable {
-      get { return targetVariable; }
-    }
+
     [Storable]
     private string[] allowedInputVariables;
@@ -128 +128 @@
       this.trainingDataset = cloner.Clone(original.trainingDataset);
       this.negativeLogLikelihood = original.negativeLogLikelihood;
-      this.targetVariable = original.targetVariable;
       this.sqrSigmaNoise = original.sqrSigmaNoise;
       if (original.meanParameter != null) {
@@ -147 +146 @@
       IEnumerable<double> hyp, IMeanFunction meanFunction, ICovarianceFunction covarianceFunction,
       bool scaleInputs = true)
-      : base() {
+      : base(targetVariable) {
       this.name = ItemName;
       this.description = ItemDescription;
       this.meanFunction = (IMeanFunction)meanFunction.Clone();
       this.covarianceFunction = (ICovarianceFunction)covarianceFunction.Clone();
-      this.targetVariable = targetVariable;
       this.allowedInputVariables = allowedInputVariables.ToArray();
 
@@ -182 +180 @@
 
       IEnumerable<double> y;
-      y = ds.GetDoubleValues(targetVariable, rows);
+      y = ds.GetDoubleValues(TargetVariable, rows);
 
       int n = x.GetLength(0);
@@ -301 +299 @@
 
     #region IRegressionModel Members
-    public IEnumerable<double> GetEstimatedValues(IDataset dataset, IEnumerable<int> rows) {
+    public override IEnumerable<double> GetEstimatedValues(IDataset dataset, IEnumerable<int> rows) {
       return GetEstimatedValuesHelper(dataset, rows);
     }
-    public GaussianProcessRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
+    public override IRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
       return new GaussianProcessRegressionSolution(this, new RegressionProblemData(problemData));
-    }
-    IRegressionSolution IRegressionModel.CreateRegressionSolution(IRegressionProblemData problemData) {
-      return CreateRegressionSolution(problemData);
     }
     #endregion
@@ -392 +387 @@
       }
     }
+
   }
 }

branches/HeuristicLab.RegressionSolutionGradientView/HeuristicLab.Algorithms.DataAnalysis/3.4/GaussianProcess/StudentTProcessModel.cs

@@ -34 +34 @@
   [StorableClass]
   [Item("StudentTProcessModel", "Represents a Student-t process posterior.")]
-  public sealed class StudentTProcessModel : NamedItem, IGaussianProcessModel {
+  public sealed class StudentTProcessModel : RegressionModel, IGaussianProcessModel {
+    public override IEnumerable<string> VariablesUsedForPrediction {
+      get { return allowedInputVariables; }
+    }
+
     [Storable]
     private double negativeLogLikelihood;
@@ -61 +65 @@
       get { return meanFunction; }
     }
-    [Storable]
-    private string targetVariable;
-    public string TargetVariable {
-      get { return targetVariable; }
-    }
+
     [Storable]
     private string[] allowedInputVariables;
@@ -131 +131 @@
       this.trainingDataset = cloner.Clone(original.trainingDataset);
       this.negativeLogLikelihood = original.negativeLogLikelihood;
-      this.targetVariable = original.targetVariable;
       if (original.meanParameter != null) {
         this.meanParameter = (double[])original.meanParameter.Clone();
@@ -151 +150 @@
       IEnumerable<double> hyp, IMeanFunction meanFunction, ICovarianceFunction covarianceFunction,
       bool scaleInputs = true)
-      : base() {
+      : base(targetVariable) {
       this.name = ItemName;
       this.description = ItemDescription;
       this.meanFunction = (IMeanFunction)meanFunction.Clone();
       this.covarianceFunction = (ICovarianceFunction)covarianceFunction.Clone();
-      this.targetVariable = targetVariable;
       this.allowedInputVariables = allowedInputVariables.ToArray();
 
@@ -186 +184 @@
 
       IEnumerable<double> y;
-      y = ds.GetDoubleValues(targetVariable, rows);
+      y = ds.GetDoubleValues(TargetVariable, rows);
 
       int n = x.GetLength(0);
@@ -318 +316 @@
 
     #region IRegressionModel Members
-    public IEnumerable<double> GetEstimatedValues(IDataset dataset, IEnumerable<int> rows) {
+    public override IEnumerable<double> GetEstimatedValues(IDataset dataset, IEnumerable<int> rows) {
       return GetEstimatedValuesHelper(dataset, rows);
     }
-    public GaussianProcessRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
+    public override IRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
       return new GaussianProcessRegressionSolution(this, new RegressionProblemData(problemData));
-    }
-    IRegressionSolution IRegressionModel.CreateRegressionSolution(IRegressionProblemData problemData) {
-      return CreateRegressionSolution(problemData);
     }
     #endregion