org.apache.ignite.ml.structures.preprocessing

Class LabeledDatasetLoader

java.lang.Object

org.apache.ignite.ml.structures.preprocessing.LabeledDatasetLoader

public class LabeledDatasetLoader
extends Object

Data pre-processing step which loads data from different file types.

Constructor Summary

Constructors

Constructor and Description
LabeledDatasetLoader()

Method Summary

Modifier and Type	Method and Description
static LabeledVectorSet	loadFromTxtFile(Path pathToFile, String separator, boolean isDistributed, boolean isFallOnBadData) Datafile should keep class labels in the first column.

Methods inherited from class java.lang.Object

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

Constructor Detail

LabeledDatasetLoader

public LabeledDatasetLoader()

Method Detail

loadFromTxtFile

public static LabeledVectorSet loadFromTxtFile(Path pathToFile,
                                               String separator,
                                               boolean isDistributed,
                                               boolean isFallOnBadData)
                                        throws IOException

Datafile should keep class labels in the first column.

Parameters:

pathToFile - Path to file.

separator - Element to tokenize row on separate tokens.

isDistributed - Generates distributed dataset if true.

isFallOnBadData - Fall on incorrect data if true.

Returns:

Labeled Dataset parsed from file.

Throws:

IOException