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

Interface ImpurityComputer<T,H extends Histogram<T,H>>

Type Parameters:

T - Base object type for histogram.

H - Type of histogram that can be used in math operations with this object.

All Superinterfaces:

Histogram<T,H>, Serializable

All Known Implementing Classes:

GiniHistogram, MSEHistogram

public interface ImpurityComputer<T,H extends Histogram<T,H>>
extends Histogram<T,H>

Interface represents an object that can compute best splitting point using features histograms.

Method Summary

Modifier and Type	Method and Description
Optional<NodeSplit>	findBestSplit() Returns best split point computed on histogram if it exists.

Methods inherited from interface org.apache.ignite.ml.dataset.feature.Histogram

addElement, buckets, getValue, isEqualTo, plus

Method Detail

findBestSplit

Optional<NodeSplit> findBestSplit()

Returns best split point computed on histogram if it exists. Split point may be absent when there is no data in histograms or split point lay in last bucket in histogram.

Returns:

Splitting point for decision tree.