Changeset 6241 for trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/RandomForest

Timestamp:

05/20/11 18:54:39 (14 years ago)

Author:

gkronber

Message:

#1473: implemented random forest wrapper for classification.

Location:

trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/RandomForest

Files:

• RandomForestClassification.cs (added)
• RandomForestClassificationSolution.cs (added)
• RandomForestModel.cs (moved) (moved from trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/RandomForest/RandomForestRegressionModel.cs) (5 diffs)
• RandomForestRegression.cs (modified) (2 diffs)
• RandomForestRegressionSolution.cs (modified) (2 diffs)

trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/RandomForest/RandomForestModel.cs

@@ -33 +33 @@
 namespace HeuristicLab.Algorithms.DataAnalysis {
   /// <summary>
-  /// Represents a random forest regression model.
+  /// Represents a random forest model for regression and classification
   /// </summary>
   [StorableClass]
-  [Item("RandomForestRegressionModel", "Represents a random forest regression model.")]
-  public sealed class RandomForestRegressionModel : NamedItem, IRandomForestRegressionModel {
+  [Item("RandomForestModel", "Represents a random forest for regression and classification.")]
+  public sealed class RandomForestModel : NamedItem, IRandomForestModel {
 
     private alglib.decisionforest randomForest;
-    /// <summary>
-    /// Gets or sets the SVM model.
-    /// </summary>
     public alglib.decisionforest RandomForest {
       get { return randomForest; }
@@ -58 +55 @@
     [Storable]
     private string[] allowedInputVariables;
-
+    [Storable]
+    private double[] classValues;
     [StorableConstructor]
-    private RandomForestRegressionModel(bool deserializing)
+    private RandomForestModel(bool deserializing)
       : base(deserializing) {
       if (deserializing)
         randomForest = new alglib.decisionforest();
     }
-    private RandomForestRegressionModel(RandomForestRegressionModel original, Cloner cloner)
+    private RandomForestModel(RandomForestModel original, Cloner cloner)
       : base(original, cloner) {
       randomForest = new alglib.decisionforest();
@@ -75 +73 @@
       targetVariable = original.targetVariable;
       allowedInputVariables = (string[])original.allowedInputVariables.Clone();
+      if (original.classValues != null)
+        this.classValues = (double[])original.classValues.Clone();
     }
-    public RandomForestRegressionModel(alglib.decisionforest randomForest, string targetVariable, IEnumerable<string> allowedInputVariables)
+    public RandomForestModel(alglib.decisionforest randomForest, string targetVariable, IEnumerable<string> allowedInputVariables, double[] classValues = null)
       : base() {
       this.name = ItemName;
@@ -83 +83 @@
       this.targetVariable = targetVariable;
       this.allowedInputVariables = allowedInputVariables.ToArray();
+      if (classValues != null)
+        this.classValues = (double[])classValues.Clone();
     }
 
     public override IDeepCloneable Clone(Cloner cloner) {
-      return new RandomForestRegressionModel(this, cloner);
+      return new RandomForestModel(this, cloner);
     }
 
@@ -103 +105 @@
         alglib.dfprocess(randomForest, x, ref y);
         yield return y[0];
+      }
+    }
+
+    public IEnumerable<double> GetEstimatedClassValues(Dataset dataset, IEnumerable<int> rows) {
+      double[,] inputData = AlglibUtil.PrepareInputMatrix(dataset, allowedInputVariables, rows);
+
+      int n = inputData.GetLength(0);
+      int columns = inputData.GetLength(1);
+      double[] x = new double[columns];
+      double[] y = new double[randomForest.innerobj.nclasses];
+
+      for (int row = 0; row < n; row++) {
+        for (int column = 0; column < columns; column++) {
+          x[column] = inputData[row, column];
+        }
+        alglib.dfprocess(randomForest, x, ref y);
+        // find class for with the largest probability value
+        int maxProbClassIndex = 0;
+        double maxProb = y[0];
+        for (int i = 1; i < y.Length; i++) {
+          if (maxProb < y[i]) {
+            maxProb = y[i];
+            maxProbClassIndex = i;
+          }
+        }
+        yield return classValues[maxProbClassIndex];
       }
     }

trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/RandomForest/RandomForestRegression.cs

@@ -87 +87 @@
       Results.Add(new Result(RandomForestRegressionModelResultName, "The random forest regression solution.", solution));
       Results.Add(new Result("Root mean square error", "The root of the mean of squared errors of the random forest regression solution on the training set.", new DoubleValue(rmsError)));
-      Results.Add(new Result("Average relative error", "The average of relative errors of the random forest regression solution on the training set.", new DoubleValue(avgRelError)));
-      Results.Add(new Result("Root mean square error (out-of-bag)", "The out-of-bag root of the mean of squared errors of the random forest regression solution on the training set.", new DoubleValue(outOfBagRmsError)));
-      Results.Add(new Result("Average relative error (out-of-bag)", "The out-of-bag average of relative errors of the random forest regression solution on the training set.", new DoubleValue(outOfBagAvgRelError)));
+      Results.Add(new Result("Average relative error", "The average of relative errors of the random forest regression solution on the training set.", new PercentValue(avgRelError)));
+      Results.Add(new Result("Root mean square error (out-of-bag)", "The out-of-bag root of the mean of squared errors of the random forest regression solution.", new DoubleValue(outOfBagRmsError)));
+      Results.Add(new Result("Average relative error (out-of-bag)", "The out-of-bag average of relative errors of the random forest regression solution.", new PercentValue(outOfBagAvgRelError)));
     }
 
@@ -116 +116 @@
       outOfBagRmsError = rep.oobrmserror;
 
-      return new RandomForestRegressionSolution(problemData, new RandomForestRegressionModel(dforest, targetVariable, allowedInputVariables));
+      return new RandomForestRegressionSolution(problemData, new RandomForestModel(dforest, targetVariable, allowedInputVariables));
     }
     #endregion

trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/RandomForest/RandomForestRegressionSolution.cs

@@ -37 +37 @@
   public sealed class RandomForestRegressionSolution : RegressionSolution, IRandomForestRegressionSolution {
 
-    public new IRandomForestRegressionModel Model {
-      get { return (IRandomForestRegressionModel)base.Model; }
+    public new IRandomForestModel Model {
+      get { return (IRandomForestModel)base.Model; }
       set { base.Model = value; }
     }
@@ -47 +47 @@
       : base(original, cloner) {
     }
-    public RandomForestRegressionSolution(IRegressionProblemData problemData, IRandomForestRegressionModel randomForestModel)
+    public RandomForestRegressionSolution(IRegressionProblemData problemData, IRandomForestModel randomForestModel)
       : base(randomForestModel, problemData) {
     }