org.apache.ignite.ml.selection.split

Class TrainTestDatasetSplitter<K,V>

java.lang.Object

org.apache.ignite.ml.selection.split.TrainTestDatasetSplitter<K,V>

Type Parameters:

K - Type of a key in upstream data.

V - Type of a value in upstream data.

All Implemented Interfaces:

Serializable

public class TrainTestDatasetSplitter<K,V>
extends Object
implements Serializable

Dataset splitter that splits dataset into train and test subsets.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
TrainTestDatasetSplitter() Constructs a new instance of train test dataset splitter.
TrainTestDatasetSplitter(UniformMapper<K,V> mapper) Constructs a new instance of train test dataset splitter.

Method Summary

Modifier and Type	Method and Description
TrainTestSplit<K,V>	split(double trainSize) Splits dataset into train and test subsets.
TrainTestSplit<K,V>	split(double trainSize, double testSize) Splits dataset into train and test subsets.

Methods inherited from class java.lang.Object

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

Constructor Detail

TrainTestDatasetSplitter

public TrainTestDatasetSplitter()

Constructs a new instance of train test dataset splitter.

TrainTestDatasetSplitter

public TrainTestDatasetSplitter(UniformMapper<K,V> mapper)

Constructs a new instance of train test dataset splitter.

Parameters:

mapper - Mapper used to map a key-value pair to a point on the segment (0, 1).

Method Detail

split

public TrainTestSplit<K,V> split(double trainSize)

Splits dataset into train and test subsets.

Parameters:

trainSize - The proportion of the dataset to include in the train split (should be between 0 and 1).

Returns:

Split with two predicates for training and testing parts.

split

public TrainTestSplit<K,V> split(double trainSize,
                                 double testSize)

Splits dataset into train and test subsets.

Parameters:

trainSize - The proportion of the dataset to include in the train split (should be between 0 and 1).

testSize - The proportion of the dataset to include in the test split (should be a number between 0 and 1).

Returns:

Split with two predicates for training and testing parts.