Changeset 17390

Timestamp:

12/20/19 13:21:43 (5 years ago)

Author:

djoedick

Message:

#3044: Added new implementation for transformation of input variables in KernelRidgeRegressionModel and added helper methods in transformation base class.

Location:

branches/3044_variableScaling

Files:

• HeuristicLab.Algorithms.DataAnalysis/3.4/KernelRidgeRegression/KernelRidgeRegressionModel.cs (modified) (3 diffs)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Transformations/Transformation.cs (modified) (2 diffs)
• HeuristicLab.Problems.DataAnalysis/3.4/Interfaces/ITransformation.cs (modified) (1 diff)

branches/3044_variableScaling/HeuristicLab.Algorithms.DataAnalysis/3.4/KernelRidgeRegression/KernelRidgeRegressionModel.cs

@@ -27 +27 @@
 using HEAL.Attic;
 using HeuristicLab.Problems.DataAnalysis;
+using HeuristicLab.Data;
 
 namespace HeuristicLab.Algorithms.DataAnalysis {
@@ -155 +156 @@
       this.kernel = (ICovarianceFunction)kernel.Clone();
       this.lambda = lambda;
-      if (scaleInputs) scaling = CreateScaling(dataset, rows, this.allowedInputVariables);
+      if (scaleInputs) {
+        var trans = new ShiftToRangeTransformation(allowedInputVariables);
+        trans.Range.Start = 0;
+        trans.Range.End = 1;
+        scaling = Transformation.CreateTransformations(trans, dataset, rows, this.allowedInputVariables);
+      } 
+
       trainX = ExtractData(dataset, rows, this.allowedInputVariables, scaling);
       var y = dataset.GetDoubleValues(targetVariable, rows).ToArray();
@@ -201 +208 @@
     }
 
-    private static ITransformation<double>[] CreateScaling(IDataset dataset, int[] rows, IReadOnlyCollection<string> allowedInputVariables) {
-      var trans = new ITransformation<double>[allowedInputVariables.Count];
-      int i = 0;
-      foreach (var variable in allowedInputVariables) {
-        var lin = new LinearTransformation(allowedInputVariables);
-        var max = dataset.GetDoubleValues(variable, rows).Max();
-        var min = dataset.GetDoubleValues(variable, rows).Min();
-        lin.Multiplier = 1.0 / (max - min);
-        lin.Addend = -min / (max - min);
-        trans[i] = lin;
-        i++;
-      }
-      return trans;
-    }
-
     private static double[,] ExtractData(IDataset dataset, IEnumerable<int> rows, IReadOnlyCollection<string> allowedInputVariables, ITransformation<double>[] scaling = null) {
       double[][] variables;

branches/3044_variableScaling/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Transformations/Transformation.cs

@@ -61 +61 @@
     protected Transformation(IEnumerable<string> allowedColumns) {
       var allowed = new ItemSet<StringValue>(allowedColumns.Select(e => new StringValue(e)));
-      Parameters.Add(new ConstrainedValueParameter<StringValue>(ColumnParameterName, "Column used for the Transformation", allowed));
+      Parameters.Add(new ConstrainedValueParameter<StringValue>(ColumnParameterName, "Column used for the Transformation", allowed, allowed.First()));
     }
+
+
+    //TODO: Fill XML comments
+    /// <summary>
+    /// Creates transformation for the provided input variables. 
+    /// </summary>
+    /// <param name="transformation"></param>
+    /// <param name="dataset"></param>
+    /// <param name="rows"></param>
+    /// <param name="allowedInputVariables"></param>
+    /// <returns></returns>
+    public static ITransformation<double>[] CreateTransformations(ITransformation<double> transformation, IDataset dataset, int[] rows, IReadOnlyCollection<string> allowedInputVariables) {
+      var trans = new ITransformation<double>[allowedInputVariables.Count];
+      int i = 0;
+      foreach (var variable in allowedInputVariables) {
+        var trans_clone = (ITransformation<double>)transformation.Clone();
+        trans_clone.Column = variable;
+        trans_clone.ConfigureParameters(dataset.GetDoubleValues(variable, rows));
+        trans[i] = trans_clone;
+        i++;
+      }
+      return trans;
+    }
+
   }
 
@@ -85 +109 @@
 
     public abstract bool Check(IEnumerable<T> data, out string errorMsg);
+
+
   }
 }

branches/3044_variableScaling/HeuristicLab.Problems.DataAnalysis/3.4/Interfaces/ITransformation.cs

@@ -28 +28 @@
   public interface ITransformation : IParameterizedItem {
     string ShortName { get; }
-    string Column { get; }
+    string Column { get; set; }
   }