Changeset 14393 for trunk/sources/HeuristicLab.Problems.DataAnalysis

Timestamp:

11/15/16 21:23:43 (8 years ago)

Author:

gkronber

Message:

#2697:

• removed AlglibUtil.cs and added an extension method .ToArray(names, rows) to IDataset instead.
• refactored transformation so that it is possible to apply an transformation without resetting the parameters
• Used transformations instead of Scaling as far as possible.
• Moved TakeEvery extension method to HL.Common

Location:

trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4

Files:

• DatasetExtensions.cs (modified) (1 diff)
• Implementation/Transformations/LinearTransformation.cs (modified) (2 diffs)
• Implementation/Transformations/ShiftStandardDistributionTransformation.cs (modified) (2 diffs)
• Implementation/Transformations/ShiftToRangeTransformation.cs (modified) (2 diffs)
• Implementation/Transformations/Transformation.cs (modified) (1 diff)
• Interfaces/ITransformation.cs (modified) (1 diff)

trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/DatasetExtensions.cs

@@ -20 +20 @@
 #endregion
 
+using System;
 using System.Collections.Generic;
+using System.Linq;
 
 namespace HeuristicLab.Problems.DataAnalysis {
   public static class DatasetExtensions {
-    public static IEnumerable<T> TakeEvery<T>(this IEnumerable<T> xs, int nth) {
-      int i = 0;
-      foreach (var x in xs) {
-        if (i % nth == 0) yield return x;
-        i++;
+    public static double[,] ToArray(this IDataset dataset, IEnumerable<string> variables, IEnumerable<int> rows) {
+      return ToArray(dataset,
+        variables,
+        transformations: variables.Select(_ => (ITransformation<double>)null),  // no transform
+        rows: rows);
+    }
+    public static double[,] ToArray(this IDataset dataset, IEnumerable<string> variables, IEnumerable<ITransformation<double>> transformations, IEnumerable<int> rows) {
+      string[] variablesArr = variables.ToArray();
+      int[] rowsArr = rows.ToArray();
+      ITransformation<double>[] transformArr = transformations.ToArray();
+      if (transformArr.Length != variablesArr.Length)
+        throw new ArgumentException("Number of variables and number of transformations must match.");
+
+      double[,] matrix = new double[rowsArr.Length, variablesArr.Length];
+
+      for (int i = 0; i < variablesArr.Length; i++) {
+        var origValues = dataset.GetDoubleValues(variablesArr[i], rowsArr);
+        var values = transformArr[i] != null ? transformArr[i].Apply(origValues) : origValues;
+        int row = 0;
+        foreach (var value in values) {
+          matrix[row, i] = value;
+          row++;
+        }
       }
+
+      return matrix;
     }
   }

trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Transformations/LinearTransformation.cs

@@ -52 +52 @@
     public double Multiplier {
       get { return MultiplierParameter.Value.Value; }
-      protected set {
+      set {
         MultiplierParameter.Value.Value = value;
       }
@@ -59 +59 @@
     public double Addend {
       get { return AddendParameter.Value.Value; }
-      protected set {
+      set {
         AddendParameter.Value.Value = value;
       }

trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Transformations/ShiftStandardDistributionTransformation.cs

@@ -71 +71 @@
 
     public override IEnumerable<double> Apply(IEnumerable<double> data) {
-      ConfigureParameters(data);
       if (OriginalStandardDeviation.IsAlmost(0.0)) {
         return data;
@@ -94 +93 @@
     }
 
-    protected void ConfigureParameters(IEnumerable<double> data) {
+    public override void ConfigureParameters(IEnumerable<double> data) {
       OriginalStandardDeviation = data.StandardDeviation();
       OriginalMean = data.Average();

trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Transformations/ShiftToRangeTransformation.cs

@@ -44 +44 @@
     }
 
-    public override IEnumerable<double> Apply(IEnumerable<double> data) {
-      ConfigureParameters(data);
-      return base.Apply(data);
-    }
-
     public override bool Check(IEnumerable<double> data, out string errorMsg) {
       ConfigureParameters(data);
@@ -54 +49 @@
     }
 
-    protected void ConfigureParameters(IEnumerable<double> data) {
+    public override void ConfigureParameters(IEnumerable<double> data) {
       double originalRangeStart = data.Min();
       double originalRangeEnd = data.Max();

trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Transformations/Transformation.cs

@@ -66 +66 @@
     protected Transformation(IEnumerable<string> allowedColumns) : base(allowedColumns) { }
 
+    public virtual void ConfigureParameters(IEnumerable<T> data) {
+      // override in transformations with parameters
+    }
+
     public abstract IEnumerable<T> Apply(IEnumerable<T> data);
+    public IEnumerable<T> ConfigureAndApply(IEnumerable<T> data) {
+      ConfigureParameters(data);
+      return Apply(data);
+    }
 
     public abstract bool Check(IEnumerable<T> data, out string errorMsg);

trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/Interfaces/ITransformation.cs

@@ -30 +30 @@
 
   public interface ITransformation<T> : ITransformation {
+    void ConfigureParameters(IEnumerable<T> data);
+    IEnumerable<T> ConfigureAndApply(IEnumerable<T> data);
     IEnumerable<T> Apply(IEnumerable<T> data);
   }