Changeset 278 for branches/3.0/sources/HeuristicLab.DataAnalysis

Timestamp:

05/30/08 13:08:42 (16 years ago)

Author:

gkronber

Message:

merged changesets r166 r168 r195:196 r209 from the trunk into the stable branch

Location:

branches/3.0/sources/HeuristicLab.DataAnalysis

Files:

• Dataset.cs (modified) (5 diffs)
• DatasetView.cs (modified) (5 diffs)

branches/3.0/sources/HeuristicLab.DataAnalysis/Dataset.cs

@@ -39 +39 @@
     private double[] samples;
     private int rows;
+    Dictionary<int, Dictionary<int, double>>[] cachedMeans;
+    Dictionary<int, Dictionary<int, double>>[] cachedRanges;
 
     public int Rows {
@@ -50 +52 @@
       set { columns = value; }
     }
-    private Dictionary<int, double[]>[] ranges;
-    private Dictionary<int, double[]>[] means;
 
     public double GetValue(int i, int j) {
@@ -94 +94 @@
       // keep a means and ranges dictionary for each column (possible target variable) of the dataset.
 
-      means = new Dictionary<int, double[]>[columns];
-      ranges = new Dictionary<int, double[]>[columns];
+      cachedMeans = new Dictionary<int, Dictionary<int, double>>[columns];
+      cachedRanges = new Dictionary<int, Dictionary<int, double>>[columns];
 
       for(int i = 0; i < columns; i++) {
-        means[i] = new Dictionary<int, double[]>();
-        ranges[i] = new Dictionary<int, double[]>();
+        cachedMeans[i] = new Dictionary<int, Dictionary<int, double>>();
+        cachedRanges[i] = new Dictionary<int, Dictionary<int, double>>();
       }
     }
@@ -200 +200 @@
     }
 
-    // return value of GetMean should be memoized because it is called repeatedly in Evaluators
     public double GetMean(int column, int from, int to) {
-      Dictionary<int, double[]> columnMeans = means[column];
-      if(columnMeans.ContainsKey(from)) {
-        double[] fromMeans = columnMeans[from];
-        if(fromMeans[to-from] >= 0.0) {
-          // already calculated
-          return fromMeans[to-from];
-        } else {
-          // not yet calculated => calculate
-          fromMeans[to-from] = CalculateMean(column, from, to);
-          return fromMeans[to-from];
-        }
+      if(!cachedMeans[column].ContainsKey(from) || !cachedMeans[column][from].ContainsKey(to)) {
+        double[] values = new double[to - from + 1];
+        for(int sample = from; sample <= to; sample++) {
+          values[sample - from] = GetValue(sample, column);
+        }
+        double mean = Statistics.Mean(values);
+        if(!cachedMeans[column].ContainsKey(from)) cachedMeans[column][from] = new Dictionary<int, double>();
+        cachedMeans[column][from][to] = mean;
+        return mean;
       } else {
-        // never saw this from-index => create a new array, initialize and recalculate for to-index
-        double[] fromMeans = new double[rows - from];
-        // fill with negative values to indicate which means have already been calculated
-        for(int i=0;i<fromMeans.Length;i++) {fromMeans[i] = -1.0;}
-        // store new array in the dictionary
-        columnMeans[from] = fromMeans;
-        // calculate for specific to-index
-        fromMeans[to-from] = CalculateMean(column, from, to);
-        return fromMeans[to-from];
-      }
-    }
-
-    private double CalculateMean(int column, int from, int to) {
-      double[] values = new double[to - from +1];
-      for(int sample = from; sample <= to; sample++) {
-        values[sample - from] = GetValue(sample, column);
-      }
-
-      return Statistics.Mean(values);
+        return cachedMeans[column][from][to];
+      }
     }
 
@@ -239 +219 @@
     }
 
-    // return value of GetRange should be memoized because it is called repeatedly in Evaluators
     public double GetRange(int column, int from, int to) {
-      Dictionary<int, double[]> columnRanges = ranges[column];
-      if(columnRanges.ContainsKey(from)) {
-        double[] fromRanges = columnRanges[from];
-        if(fromRanges[to-from] >= 0.0) {
-          // already calculated
-          return fromRanges[to-from];
-        } else {
-          // not yet calculated => calculate
-          fromRanges[to-from] = CalculateRange(column, from, to);
-          return fromRanges[to-from];
-        }
+      if(!cachedRanges[column].ContainsKey(from) || !cachedRanges[column][from].ContainsKey(to)) {
+        double[] values = new double[to - from + 1];
+        for(int sample = from; sample <= to; sample++) {
+          values[sample - from] = GetValue(sample, column);
+        }
+        double range = Statistics.Range(values);
+        if(!cachedRanges[column].ContainsKey(from)) cachedRanges[column][from] = new Dictionary<int, double>();
+        cachedRanges[column][from][to] = range;
+        return range;
       } else {
-        // never saw this from-index => create a new array, initialize and recalculate for to-index
-        double[] fromRanges = new double[rows - from];
-        // fill with negative values to indicate which means have already been calculated
-        for(int i = 0; i < fromRanges.Length; i++) { fromRanges[i] = -1.0; }
-        // store in dictionary
-        columnRanges[from] = fromRanges;
-        // calculate for specific to-index
-        fromRanges[to-from] = CalculateRange(column, from, to);
-        return fromRanges[to-from];
-      }
-    }
-
-    private double CalculateRange(int column, int from, int to) {
-      double[] values = new double[to - from + 1];
-      for(int sample = from; sample <= to; sample++) {
-        values[sample - from] = GetValue(sample, column);
-      }
-
-      return Statistics.Range(values);
+        return cachedRanges[column][from][to];
+      }
     }
   }

branches/3.0/sources/HeuristicLab.DataAnalysis/DatasetView.cs

@@ -33 +33 @@
       set {
         Item = value;
-        Item.Changed += new EventHandler(Item_Changed);
+        Refresh();
       }
     }
 
-    void Item_Changed(object sender, EventArgs e) {
-      Refresh();
-    }
-
-    public DatasetView() : base() {
+    public DatasetView()
+      : base() {
       InitializeComponent();
       openFileDialog = new OpenFileDialog();
     }
 
-    public DatasetView(Dataset dataset) :this() {
+    public DatasetView(Dataset dataset)
+      : this() {
       this.Dataset = dataset;
     }
@@ -52 +50 @@
     protected override void UpdateControls() {
       base.UpdateControls();
-      if(Dataset != null) {
+      if (Dataset != null) {
         int rows = Dataset.Rows;
         int columns = Dataset.Columns;
-
         nameTextBox.Text = Dataset.Name;
         rowsTextBox.Text = rows + "";
@@ -61 +58 @@
         dataGridView.ColumnCount = columns;
         dataGridView.RowCount = rows;
-        for(int i = 0; i < rows; i++) {
-          for(int j = 0; j < columns; j++) {
+        for (int i = 0; i < rows; i++) {
+          for (int j = 0; j < columns; j++) {
             dataGridView.Rows[i].Cells[j].Value = Dataset.GetValue(i, j);
           }
@@ -85 +82 @@
 
     private void dataGridView_CellValidating(object sender, DataGridViewCellValidatingEventArgs e) {
-      if(ValidateData((string)e.FormattedValue)) {
+      if (ValidateData((string)e.FormattedValue)) {
         SetArrayElement(e.RowIndex, e.ColumnIndex, (string)e.FormattedValue);
         e.Cancel = false;
@@ -97 +94 @@
       double result;
       double.TryParse(element, out result);
-      if(result != Dataset.GetValue(row, column)) {
+      if (result != Dataset.GetValue(row, column)) {
         Dataset.SetValue(row, column, result);
         Dataset.FireChanged();