com.datumbox.framework.core.machinelearning.modelselection.splitters

Class ShuffleSplitter

java.lang.Object

com.datumbox.framework.core.machinelearning.common.abstracts.modelselection.AbstractSplitter

com.datumbox.framework.core.machinelearning.modelselection.splitters.ShuffleSplitter

public class ShuffleSplitter
extends AbstractSplitter

The ShuffleSplitter partitions the data in train/test splits.

Nested Class Summary

Nested classes/interfaces inherited from class com.datumbox.framework.core.machinelearning.common.abstracts.modelselection.AbstractSplitter

AbstractSplitter.Split

Constructor Summary

Constructors

Constructor and Description
ShuffleSplitter(double proportion, int splits) The default constructor which receives the train size and the number of splits.
ShuffleSplitter(double proportion, int splits, java.util.Random random) The auxiliary constructor which receives the train size, the number of splits and a random generator.

Method Summary

Modifier and Type	Method and Description
java.util.Iterator<AbstractSplitter.Split>	split(Dataframe dataset) Produces a series of train/test splits on the provided dataframe.

Methods inherited from class java.lang.Object

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

Constructor Detail

ShuffleSplitter

public ShuffleSplitter(double proportion,
                       int splits)

The default constructor which receives the train size and the number of splits.

Parameters:

proportion: - the proportion of the dataset in the train split

splits: - number of reshuffling and splitting iterations

ShuffleSplitter

public ShuffleSplitter(double proportion,
                       int splits,
                       java.util.Random random)

The auxiliary constructor which receives the train size, the number of splits and a random generator.

Parameters:

proportion: - the proportion of the dataset in the train split

splits: - number of reshuffling and splitting iterations

Method Detail

split

public java.util.Iterator<AbstractSplitter.Split> split(Dataframe dataset)

Description copied from class: AbstractSplitter

Produces a series of train/test splits on the provided dataframe.

Specified by:

split in class AbstractSplitter

Returns: