org.apache.spark.mllib.util

Class DataValidators

java.lang.Object

org.apache.spark.mllib.util.DataValidators

All Implemented Interfaces:

Logging

public class DataValidators
extends java.lang.Object
implements Logging

:: DeveloperApi :: A collection of methods used to validate data before applying ML algorithms.

Constructor Summary

Constructors

Constructor and Description
DataValidators()

Method Summary

Methods

Modifier and Type	Method and Description
static scala.Function1<RDD<LabeledPoint>,java.lang.Object>	binaryLabelValidator() Function to check if labels used for classification are either zero or one.
static scala.Function1<RDD<LabeledPoint>,java.lang.Object>	multiLabelValidator(int k) Function to check if labels used for k class multi-label classification are in the range of {0, 1, ..., k - 1}.

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.spark.Logging

initializeIfNecessary, initializeLogging, isTraceEnabled, log_, log, logDebug, logDebug, logError, logError, logInfo, logInfo, logName, logTrace, logTrace, logWarning, logWarning

Constructor Detail

DataValidators

public DataValidators()

Method Detail

binaryLabelValidator

public static scala.Function1<RDD<LabeledPoint>,java.lang.Object> binaryLabelValidator()

Function to check if labels used for classification are either zero or one.

Returns:

True if labels are all zero or one, false otherwise.

multiLabelValidator

public static scala.Function1<RDD<LabeledPoint>,java.lang.Object> multiLabelValidator(int k)

Function to check if labels used for k class multi-label classification are in the range of {0, 1, ..., k - 1}.

Parameters:

k - (undocumented)

Returns:

True if labels are all in the range of {0, 1, ..., k-1}, false otherwise.