org.apache.spark.ml.tree

Class InternalNode

java.lang.Object

org.apache.spark.ml.tree.Node

org.apache.spark.ml.tree.InternalNode

All Implemented Interfaces:

java.io.Serializable

public final class InternalNode
extends Node

:: DeveloperApi :: Internal Decision Tree node. param: prediction Prediction this node would make if it were a leaf node param: impurity Impurity measure at this node (for training data) param: gain Information gain value. Values < 0 indicate missing values; this quirk will be removed with future updates. param: leftChild Left-hand child node param: rightChild Right-hand child node param: split Information about the test used to split to the left or right child.

See Also:

Serialized Form

Method Summary

Methods

Modifier and Type	Method and Description
double	gain()
double	impurity() Impurity measure at this node (for training data)
org.apache.spark.mllib.tree.impurity.ImpurityCalculator	impurityStats() Statistics aggregated from training data at this node, used to compute prediction, impurity, and probabilities.
Node	leftChild()
double	prediction() Prediction a leaf node makes, or which an internal node would make if it were a leaf node
Node	rightChild()
Split	split()
java.lang.String	toString()

Methods inherited from class org.apache.spark.ml.tree.Node

fromOld

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Method Detail

prediction

public double prediction()

Description copied from class: Node

Prediction a leaf node makes, or which an internal node would make if it were a leaf node

Specified by:

prediction in class Node

impurity

public double impurity()

Description copied from class: Node

Impurity measure at this node (for training data)

Specified by:

impurity in class Node

gain

public double gain()

leftChild

public Node leftChild()

rightChild

public Node rightChild()

split

public Split split()

impurityStats

public org.apache.spark.mllib.tree.impurity.ImpurityCalculator impurityStats()

Description copied from class: Node

Statistics aggregated from training data at this node, used to compute prediction, impurity, and probabilities. For classification, the array of class counts must be normalized to a probability distribution.

Returns:

(undocumented)

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object