Package org.apache.spark.ml.tree

Interface TreeEnsembleModel<M extends DecisionTreeModel>

Type Parameters:
M - Type of tree model in this ensemble
All Known Implementing Classes:
GBTClassificationModel, GBTRegressionModel, RandomForestClassificationModel, RandomForestRegressionModel
public interface TreeEnsembleModel<M extends DecisionTreeModel>
Abstraction for models which are ensembles of decision trees

  • Method Details

    • trees

      M[] trees()
      Trees in this ensemble. Warning: These have null parent Estimators.

    • getTree

      M getTree(int i)

    • treeWeights

      double[] treeWeights()
      Weights for each tree, zippable with trees()

    • javaTreeWeights

      Vector javaTreeWeights()
      Weights used by the python wrappers.

    • toString

      String toString()
      Summary of the model
      Overrides:
      toString in class Object

    • toDebugString

      String toDebugString()
      Full description of model

    • totalNumNodes

      int totalNumNodes()
      Total number of nodes, summed over all trees in the ensemble.

    • predictLeaf

      Vector predictLeaf(Vector features)
      Parameters:
      features - (undocumented)
      Returns:
      The indices of the leaves corresponding to the feature vector. Leaves are indexed in pre-order from 0.

    • getLeafField

      StructField getLeafField(String leafCol)

    • getEstimatedSize

      long getEstimatedSize()