org.apache.ignite.ml.tree.randomforest.data.impurity

Class ImpurityHistogramsComputer.NodeImpurityHistograms<S extends ImpurityComputer<BootstrappedVector,S>>

java.lang.Object

org.apache.ignite.ml.tree.randomforest.data.impurity.ImpurityHistogramsComputer.NodeImpurityHistograms<S>

All Implemented Interfaces:

Serializable

Enclosing class:

ImpurityHistogramsComputer<S extends ImpurityComputer<BootstrappedVector,S>>

public static class ImpurityHistogramsComputer.NodeImpurityHistograms<S extends ImpurityComputer<BootstrappedVector,S>>
extends Object
implements Serializable

Class represents per feature statistics for impurity computing.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
NodeImpurityHistograms(NodeId nodeId) Create an instance of NodeImpurityHistograms.

Method Summary

Modifier and Type	Method and Description
Optional<NodeSplit>	findBestSplit() Find best split point, based on feature statistics.
NodeId	getNodeId()
ImpurityHistogramsComputer.NodeImpurityHistograms<S>	plus(ImpurityHistogramsComputer.NodeImpurityHistograms<S> other) Store features statistics from other instance.

Methods inherited from class java.lang.Object

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

Constructor Detail

NodeImpurityHistograms

public NodeImpurityHistograms(NodeId nodeId)

Create an instance of NodeImpurityHistograms.

Parameters:

nodeId - Node id.

Method Detail

plus

public ImpurityHistogramsComputer.NodeImpurityHistograms<S> plus(ImpurityHistogramsComputer.NodeImpurityHistograms<S> other)

Store features statistics from other instance.

Parameters:

other - Other instance.

getNodeId

public NodeId getNodeId()

findBestSplit

public Optional<NodeSplit> findBestSplit()

Find best split point, based on feature statistics.

Returns:

Best split point if it exists.