Class DecisionTreeModelReadWrite.NodeData

Object
org.apache.spark.ml.tree.DecisionTreeModelReadWrite.NodeData
All Implemented Interfaces:
Serializable, scala.Equals, scala.Product, scala.Serializable
Enclosing class:
DecisionTreeModelReadWrite

public static class DecisionTreeModelReadWrite.NodeData extends Object implements scala.Product, scala.Serializable
Info for a Node

param: id Index used for tree reconstruction. Indices follow a pre-order traversal. param: impurityStats Stats array. Impurity type is stored in metadata. param: rawCount The unweighted number of samples falling in this node. param: gain Gain, or arbitrary value if leaf node. param: leftChild Left child index, or arbitrary value if leaf node. param: rightChild Right child index, or arbitrary value if leaf node. param: split Split info, or arbitrary value if leaf node.

See Also:
  • Constructor Details

    • NodeData

      public NodeData(int id, double prediction, double impurity, double[] impurityStats, long rawCount, double gain, int leftChild, int rightChild, DecisionTreeModelReadWrite.SplitData split)
  • Method Details

    • id

      public int id()
    • prediction

      public double prediction()
    • impurity

      public double impurity()
    • impurityStats

      public double[] impurityStats()
    • rawCount

      public long rawCount()
    • gain

      public double gain()
    • leftChild

      public int leftChild()
    • rightChild

      public int rightChild()
    • split