org.apache.ignite.ml.tree

Class DecisionTreeConditionalNode

java.lang.Object

org.apache.ignite.ml.tree.DecisionTreeConditionalNode

All Implemented Interfaces:

Serializable, AutoCloseable, IgniteModel<Vector,Double>, Model<Vector,Double>, DecisionTreeNode

public final class DecisionTreeConditionalNode
extends Object
implements DecisionTreeNode

Decision tree conditional (non-leaf) node.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
DecisionTreeConditionalNode(int col, double threshold, DecisionTreeNode thenNode, DecisionTreeNode elseNode, DecisionTreeNode missingNode) Constructs a new instance of decision tree conditional node.

Method Summary

Modifier and Type	Method and Description
int	getCol()
DecisionTreeNode	getElseNode()
DecisionTreeNode	getMissingNode()
DecisionTreeNode	getThenNode()
double	getThreshold()
Double	predict(Vector features) Make a prediction for the specified input arguments.
void	setElseNode(DecisionTreeNode elseNode)
void	setMissingNode(DecisionTreeNode missingNode)
void	setThenNode(DecisionTreeNode thenNode)
String	toString()
String	toString(boolean pretty)

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.ignite.ml.IgniteModel

andBefore, andThen, andThen, close, combine

Constructor Detail

DecisionTreeConditionalNode

public DecisionTreeConditionalNode(int col,
                                   double threshold,
                                   DecisionTreeNode thenNode,
                                   DecisionTreeNode elseNode,
                                   DecisionTreeNode missingNode)

Constructs a new instance of decision tree conditional node.

Parameters:

col - Column of the value to be tested.

threshold - Threshold.

thenNode - Node that will be used in case tested value is greater then threshold.

elseNode - Node that will be used in case tested value is not greater then threshold.

missingNode - Node that will be used in case tested value is not presented.

Method Detail

predict

public Double predict(Vector features)

Make a prediction for the specified input arguments.

Specified by:

predict in interface Model<Vector,Double>

Parameters:

features - Input arguments.

Returns:

Prediction result.

getCol

public int getCol()

getThreshold

public double getThreshold()

getThenNode

public DecisionTreeNode getThenNode()

setThenNode

public void setThenNode(DecisionTreeNode thenNode)

getElseNode

public DecisionTreeNode getElseNode()

setElseNode

public void setElseNode(DecisionTreeNode elseNode)

getMissingNode

public DecisionTreeNode getMissingNode()

setMissingNode

public void setMissingNode(DecisionTreeNode missingNode)

toString

public String toString()

Overrides:

toString in class Object

toString

public String toString(boolean pretty)

Specified by:

toString in interface IgniteModel<Vector,Double>

Parameters:

pretty - Use pretty mode.