Changeset 14015 for trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees

Timestamp:

07/07/16 12:51:31 (8 years ago)

Author:

gkronber

Message:

#2612: added NaN handling for the evaluation of regression tree models (GBT)

File:

• trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees/RegressionTreeModel.cs (modified) (1 diff)

trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees/RegressionTreeModel.cs

@@ -170 +170 @@
         if (node.VarName == TreeNode.NO_VARIABLE)
           return node.Val;
-        if (columnCache[nodeIdx] == null) {
+        if (columnCache[nodeIdx] == null || double.IsNaN(columnCache[nodeIdx][row])) {
           if (node.WeightLeft.IsAlmost(-1.0)) throw new InvalidOperationException("Cannot calculate partial dependence for trees loaded from older versions of HeuristicLab.");
           // weighted average for partial dependence plot (recursive here because we need to calculate both sub-trees)