Changeset 12711 for stable/HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees

Timestamp:

07/10/15 13:44:32 (9 years ago)

Author:

gkronber

Message:

#2261: merged r12700 and r12710 from trunk to stable

Location:

stable

Files:

• . (modified) (1 prop)
• HeuristicLab.Algorithms.DataAnalysis (modified) (1 prop)
• HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees (copied) (copied from trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees)
• HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees/GradientBoostedTreesAlgorithmStatic.cs (modified) (7 diffs)

stable/HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees/GradientBoostedTreesAlgorithmStatic.cs

@@ -48 +48 @@
       internal double r { get; private set; }
       internal double m { get; private set; }
+      internal int[] trainingRows { get; private set; }
+      internal int[] testRows { get; private set; }
       internal RegressionTreeBuilder treeBuilder { get; private set; }
 
@@ -71 +73 @@
         random = new MersenneTwister(randSeed);
         this.problemData = problemData;
+        this.trainingRows = problemData.TrainingIndices.ToArray();
+        this.testRows = problemData.TestIndices.ToArray();
         this.lossFunction = lossFunction;
 
-        int nRows = problemData.TrainingIndices.Count();
+        int nRows = trainingRows.Length;
 
-        y = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, problemData.TrainingIndices).ToArray();
+        y = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, trainingRows).ToArray();
 
         treeBuilder = new RegressionTreeBuilder(problemData, random);
@@ -84 +88 @@
         double f0 = lossFunction.LineSearch(y, zeros, activeIdx, 0, nRows - 1); // initial constant value (mean for squared errors)
         pred = Enumerable.Repeat(f0, nRows).ToArray();
-        predTest = Enumerable.Repeat(f0, problemData.TestIndices.Count()).ToArray();
+        predTest = Enumerable.Repeat(f0, testRows.Length).ToArray();
         pseudoRes = new double[nRows];
 
@@ -106 +110 @@
       }
       public double GetTestLoss() {
-        var yTest = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, problemData.TestIndices);
-        var nRows = problemData.TestIndices.Count();
+        var yTest = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, testRows);
+        var nRows = testRows.Length;
         return lossFunction.GetLoss(yTest, predTest) / nRows;
       }
@@ -160 +164 @@
       var activeIdx = gbmState.activeIdx;
       var pseudoRes = gbmState.pseudoRes;
+      var trainingRows = gbmState.trainingRows;
+      var testRows = gbmState.testRows;
 
       // copy output of gradient function to pre-allocated rim array (pseudo-residual per row and model)
@@ -170 +176 @@
 
       int i = 0;
-      foreach (var pred in tree.GetEstimatedValues(problemData.Dataset, problemData.TrainingIndices)) {
+      foreach (var pred in tree.GetEstimatedValues(problemData.Dataset, trainingRows)) {
         yPred[i] = yPred[i] + nu * pred;
         i++;
@@ -176 +182 @@
       // update predictions for validation set
       i = 0;
-      foreach (var pred in tree.GetEstimatedValues(problemData.Dataset, problemData.TestIndices)) {
+      foreach (var pred in tree.GetEstimatedValues(problemData.Dataset, testRows)) {
         yPredTest[i] = yPredTest[i] + nu * pred;
         i++;