it.uniroma2.sag.kelp.data.example

Class Example

java.lang.Object

it.uniroma2.sag.kelp.data.example.Example

All Implemented Interfaces:

Serializable

Direct Known Subclasses:

SimpleExample

public abstract class Example
extends Object
implements Serializable

It is the instance of an example in the Machine Learning context. An Example consists of one or more representation of the same object and zero (unsupervised learning) or more classificationLabels (supervised learning. Single or multitask learning)

Author:

Simone Filice

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
Example() Initializes an empty example (0 classificationLabels and 0 regression values)

Method Summary

Methods

Modifier and Type	Method and Description
void	addLabel(Label label) Adds a label to the example
abstract void	addRepresentation(String representationName, Representation representation) Adds a representation to this example
Example	duplicate()
boolean	equals(Object example)
HashSet<Label>	getClassificationLabels()
long	getId() Returns a unique identifier of the example.
Label[]	getLabels() Returns the classification classificationLabels of this example
int	getNumberOfClassificationLabels() Returns the number of classification classificationLabels whose this instance is a positive example
int	getNumberOfRegressionLabels() Returns the number of regression classificationLabels
abstract int	getNumberOfRepresentations() Returns the number of representations in which this example is modeled
NumericLabel[]	getRegressionLabels() Returns the classificationLabels of this example
Float	getRegressionValue(Label property) Returns the numeric value associated to a label
ArrayList<NumericLabel>	getRegressionValues()
abstract Representation	getRepresentation(String representationName) Returns the representation corresponding to representationName
abstract Map<String,Representation>	getRepresentations() Returns the example representations
protected String	getTextualLabelPart()
abstract boolean	isCompatible(Example example) Evaluates whether an example is compatible with this one, i.e., they have the same structure in terms of Example type and representations
boolean	isExampleOf(Label label) Asserts whether this is a positive example for the input label or not
abstract void	manipulate(Manipulator manipulator) Manipulate this example accordingly to the provided manipulator
void	setClassificationLabels(HashSet<Label> labels)
void	setLabels(Label[] labels) Sets the example classificationLabels
void	setRegressionValues(ArrayList<NumericLabel> regressionValues)
abstract void	setRepresentations(HashMap<String,Representation> representations) Sets the example representations

Methods inherited from class java.lang.Object

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

Constructor Detail

Example

public Example()

Initializes an empty example (0 classificationLabels and 0 regression values)

Method Detail

getId

public long getId()

Returns a unique identifier of the example. The Unique identifier is a number univocally and statically assigned to the example during its initialization.

Note: initializing two identical examples will make them having two different IDs

Returns:

the example ID

setLabels

public void setLabels(Label[] labels)

Sets the example classificationLabels

Parameters:

classificationLabels - of which this instance is a positive example

getLabels

public Label[] getLabels()

Returns the classification classificationLabels of this example

Returns:

the classification classificationLabels of this example

getRegressionLabels

public NumericLabel[] getRegressionLabels()

Returns the classificationLabels of this example

Returns:

the classificationLabels of this example

addLabel

public void addLabel(Label label)

Adds a label to the example

Parameters:

label -

getNumberOfClassificationLabels

public int getNumberOfClassificationLabels()

Returns the number of classification classificationLabels whose this instance is a positive example

Returns:

the number of classificationLabels whose this instance is a positive example

getNumberOfRegressionLabels

public int getNumberOfRegressionLabels()

Returns the number of regression classificationLabels

Returns:

the number of regression classificationLabels

isExampleOf

public boolean isExampleOf(Label label)

Asserts whether this is a positive example for the input label or not

Parameters:

label -

Returns:

true if this is a positive example of the class label, false otherwise

getRegressionValue

public Float getRegressionValue(Label property)

Returns the numeric value associated to a label

Parameters:

propertyName - the regression label identifier

Returns:

the numeric value associated to the label identified by propertyName or null if that label is not a NumericLabel or this example does not contain that label

getClassificationLabels

public HashSet<Label> getClassificationLabels()

Returns:

the classificationLabels

setClassificationLabels

public void setClassificationLabels(HashSet<Label> labels)

Parameters:

classificationLabels - the classificationLabels to set

getRegressionValues

public ArrayList<NumericLabel> getRegressionValues()

Returns:

the regressionValues

setRegressionValues

public void setRegressionValues(ArrayList<NumericLabel> regressionValues)

Parameters:

regressionValues - the regressionValues to set

equals

public boolean equals(Object example)

Overrides:

equals in class Object

duplicate

public Example duplicate()

Returns:

A copu of the current object with a different identifier, i.e. the exampleId

getTextualLabelPart

protected String getTextualLabelPart()

setRepresentations

public abstract void setRepresentations(HashMap<String,Representation> representations)

Sets the example representations

Parameters:

representations - to associate to this example

getRepresentations

public abstract Map<String,Representation> getRepresentations()

Returns the example representations

Returns:

the representations of this example

addRepresentation

public abstract void addRepresentation(String representationName,
                     Representation representation)

Adds a representation to this example

Parameters:

representationName - the identifier of the representation to be added

representation - the representation to be added

getNumberOfRepresentations

public abstract int getNumberOfRepresentations()

Returns the number of representations in which this example is modeled

Returns:

the number of representations

getRepresentation

public abstract Representation getRepresentation(String representationName)

Returns the representation corresponding to representationName

Parameters:

representationName - it is a representation identifier

Returns:

the representation corresponding to representationName

manipulate

public abstract void manipulate(Manipulator manipulator)

Manipulate this example accordingly to the provided manipulator

Parameters:

manipulator - the manipulator

isCompatible

public abstract boolean isCompatible(Example example)

Evaluates whether an example is compatible with this one, i.e., they have the same structure in terms of Example type and representations

Parameters:

example - the example to be compared with this one

Returns:

whether this example is compatible with the input one