org.apache.mahout.classifier.df.data

Class Dataset

java.lang.Object

org.apache.mahout.classifier.df.data.Dataset

public class Dataset
extends Object

Contains information about the attributes.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	Dataset.Attribute Attributes type

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	Dataset()

Method Summary

Methods

Modifier and Type	Method and Description
boolean	equals(Object obj)
static Dataset	fromJSON(String json) De-serialize an instance from a string
Dataset.Attribute	getAttribute(int attr)
int[]	getIgnored()
double	getLabel(Instance instance)
int	getLabelId()
String	getLabelString(double code) Returns the label value in the data This method can be used when the criterion variable is the categorical attribute.
int	hashCode()
boolean	isNumerical(int attr) Is this a numerical attribute ?
int	labelCode(String label) Returns the code used to represent the label value in the data
String[]	labels()
static Dataset	load(org.apache.hadoop.conf.Configuration conf, org.apache.hadoop.fs.Path path) Loads the dataset from a file
int	nbAttributes()
int	nblabels()
int	nbValues(int attr)
String	toJSON() Serialize this instance to JSON
String	toString()
int	valueOf(int attr, String token) Converts a token to its corresponding integer code for a given attribute

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Detail

Dataset

protected Dataset()

Method Detail

nbValues

public int nbValues(int attr)

labels

public String[] labels()

nblabels

public int nblabels()

getLabelId

public int getLabelId()

getLabel

public double getLabel(Instance instance)

getAttribute

public Dataset.Attribute getAttribute(int attr)

labelCode

public int labelCode(String label)

Returns the code used to represent the label value in the data

Parameters:

label - label's value to code

Returns:

label's code

getLabelString

public String getLabelString(double code)

Returns the label value in the data This method can be used when the criterion variable is the categorical attribute.

Parameters:

code - label's code

Returns:

label's value

toString

public String toString()

Overrides:

toString in class Object

valueOf

public int valueOf(int attr,
          String token)

Converts a token to its corresponding integer code for a given attribute

Parameters:

attr - attribute index

getIgnored

public int[] getIgnored()

nbAttributes

public int nbAttributes()

Returns:

number of attributes

isNumerical

public boolean isNumerical(int attr)

Is this a numerical attribute ?

Parameters:

attr - index of the attribute to check

Returns:

true if the attribute is numerical

equals

public boolean equals(Object obj)

Overrides:

equals in class Object

hashCode

public int hashCode()

Overrides:

hashCode in class Object

load

public static Dataset load(org.apache.hadoop.conf.Configuration conf,
           org.apache.hadoop.fs.Path path)
                    throws IOException

Loads the dataset from a file

Throws:

IOException

toJSON

public String toJSON()

Serialize this instance to JSON

Returns:

some JSON

fromJSON

public static Dataset fromJSON(String json)

De-serialize an instance from a string

Parameters:

json - From which an instance is created

Returns:

A shiny new Dataset