Changeset 5931 for branches/QAP/HeuristicLab.Analysis/3.3/MultidimensionalScaling

Timestamp:

04/02/11 16:45:04 (14 years ago)

Author:

abeham

Message:

#1330

• changes according to the review

File:

• branches/QAP/HeuristicLab.Analysis/3.3/MultidimensionalScaling/MultidimensionalScaling.cs (modified) (8 diffs)

branches/QAP/HeuristicLab.Analysis/3.3/MultidimensionalScaling/MultidimensionalScaling.cs

@@ -29 +29 @@
     /// Performs the Kruskal-Shepard algorithm and applies a gradient descent method
     /// to fit the coordinates such that the difference between the fit distances
-    /// and the actual distances becomes minimal.
+    /// and the dissimilarities becomes minimal.
     /// </summary>
     /// <remarks>
     /// It will initialize the coordinates in a deterministic fashion such that all initial points are equally spaced on a circle.
     /// </remarks>
-    /// <param name="dissimilarities">A symmetric NxN matrix that specifies the distances between each element i and j. Diagonal elements are ignored.</param>
+    /// <param name="dissimilarities">A symmetric NxN matrix that specifies the dissimilarities between each element i and j. Diagonal elements are ignored.</param>
     /// 
     /// <returns>A Nx2 matrix where the first column represents the x- and the second column the y coordinates.</returns>
@@ -43 +43 @@
       int dimension = dissimilarities.Rows;
       if (dimension == 1) return new DoubleMatrix(new double[,] { { 0, 0 } });
-      else if (dimension == 2) return new DoubleMatrix(new double[,] { { 0, dissimilarities[0, 1] } });
+      else if (dimension == 2) return new DoubleMatrix(new double[,] { { 0, 0 }, { 0, dissimilarities[0, 1] } });
 
       DoubleMatrix coordinates = new DoubleMatrix(dimension, 2);
@@ -58 +58 @@
     /// Performs the Kruskal-Shepard algorithm and applies a gradient descent method
     /// to fit the coordinates such that the difference between the fit distances
-    /// and the actual distances is minimal.
+    /// and the dissimilarities is minimal.
     /// </summary>
     /// <remarks>
     /// It will use a pre-initialized x,y-coordinates matrix as a starting point of the gradient descent.
     /// </remarks>
-    /// <param name="dissimilarities">A symmetric NxN matrix that specifies the distances between each element i and j. Diagonal elements are ignored.</param>
-    /// 
+    /// <param name="dissimilarities">A symmetric NxN matrix that specifies the dissimilarities between each element i and j. Diagonal elements are ignored.</param>
     /// <param name="coordinates">The Nx2 matrix of initial coordinates.</param>
+    /// <param name="maximumIterations">The number of iterations for which the algorithm should run.
+    /// In every iteration it tries to find the best location for every item.</param>
     /// <returns>A Nx2 matrix where the first column represents the x- and the second column the y coordinates.</returns>
-    public static DoubleMatrix KruskalShepard(DoubleMatrix dissimilarities, DoubleMatrix coordinates) {
+    public static DoubleMatrix KruskalShepard(DoubleMatrix dissimilarities, DoubleMatrix coordinates, int maximumIterations = 100) {
       int dimension = dissimilarities.Rows;
       if (dimension != dissimilarities.Columns || coordinates.Rows != dimension) throw new ArgumentException("The number of coordinates and the number of rows and columns in the dissimilarities matrix do not match.");
@@ -78 +79 @@
       alglib.mincgreport rep;
 
-      for (int iterations = 0; iterations < 10; iterations++) {
+      for (int iterations = 0; iterations < maximumIterations; iterations++) {
+        bool changed = false;
         for (int i = 0; i < dimension; i++) {
           double[] c = new double[] { coordinates[i, 0], coordinates[i, 1] };
@@ -93 +95 @@
           } catch (alglib.alglibexception) { }
           if (!double.IsNaN(c[0]) && !double.IsNaN(c[1])) {
+            changed = changed || (coordinates[i, 0] != c[0]) || (coordinates[i, 1] != c[1]);
             coordinates[i, 0] = c[0];
             coordinates[i, 1] = c[1];
           }
         }
+        if (!changed) break;
       }
       return coordinates;
@@ -106 +110 @@
       Info info = (obj as Info);
       for (int i = 0; i < info.Coordinates.Rows; i++) {
-        double c = info.Distances[info.Row, i];
+        double c = info.Dissimilarities[info.Row, i];
         if (i != info.Row) {
           double a = info.Coordinates[i, 0];
@@ -143 +147 @@
       for (int i = 0; i < dimension - 1; i++) {
         for (int j = i + 1; j < dimension; j++) {
-          if (dissimilarities[i, j] != dissimilarities[j, i]) throw new ArgumentException("Distances is not a symmetric matrix.", "distances");
+          if (dissimilarities[i, j] != dissimilarities[j, i]) throw new ArgumentException("Dissimilarities is not a symmetric matrix.", "dissimilarities");
           if (dissimilarities[i, j] != 0) {
             stress += Stress(coordinates[i, 0], coordinates[i, 1], dissimilarities[i, j], coordinates[j, 0], coordinates[j, 1]);
@@ -155 +159 @@
     private class Info {
       public DoubleMatrix Coordinates { get; set; }
-      public DoubleMatrix Distances { get; set; }
+      public DoubleMatrix Dissimilarities { get; set; }
       public int Row { get; set; }
 
       public Info(DoubleMatrix c, DoubleMatrix d, int r) {
         Coordinates = c;
-        Distances = d;
+        Dissimilarities = d;
         Row = r;
       }