Package com.jml.preprocessing

Class DataSplitter

java.lang.Object

com.jml.preprocessing.DataSplitter

public class DataSplitter
extends Object

A class that provides a method for splitting a dataset into a training and testing dataset.

Method Summary

Modifier and Type	Method	Description
static Map<String,double[][]>	trainTestSplit(double[][] X, double[][] y, double testSize)	Splits a dataset with features and targets randomly into disjoint training and testing datasets.

Methods inherited from class java.lang.Object

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

Method Details

trainTestSplit

public static Map<String,double[][]> trainTestSplit(double[][] X,
 double[][] y,
 double testSize)

Splits a dataset with features and targets randomly into disjoint training and testing datasets. The arrays are first shuffled using the Fisher–Yates algorithm such that both X and y are shuffled the same. Then the arrays are split into a training and testing dataset.

Parameters:

X - Features of the dataset.

y - Targets of the dataset.

testSize - Percent of data to include in the testing dataset. Should be a value between 0 and 1 (inclusive).

Returns:

A hashmap containing the split dataset. Use the following keys to obtain the split data:

     key: "xTrain" -> X training data.
     key: "yTrain" -> y training data.
     key: "xTest" -> X testing data.
     key: "yTest" -> y testing data.