jsat.classifiers.linear

Class SPA

java.lang.Object

jsat.classifiers.BaseUpdateableClassifier

jsat.classifiers.linear.SPA

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, Classifier, UpdateableClassifier, Parameterized, SimpleWeightVectorModel

public class SPA
extends BaseUpdateableClassifier
implements Parameterized, SimpleWeightVectorModel

Support class Passive Aggressive (SPA) is a multi class generalization of PassiveAggressive. It works in the same philosophy, and can obtain better multi class accuracy then PA used with a meta learner.
SPA is more sensitive to small values for the aggressiveness parameter.
If working with a binary classification problem, SPA reduces to PA, and the original PA implementation should be used instead.
By default, the biast term is not used.

See:
Matsushima, S., Shimizu, N., Yoshida, K., Ninomiya, T.,&Nakagawa, H. (2010). Exact Passive-Aggressive Algorithm for Multiclass Classification Using Support Class. SIAM International Conference on Data Mining - SDM (pp. 303–314). Retrieved from here

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
SPA() Creates a new Passive Aggressive learner that does 10 epochs and uses PA2.
SPA(int epochs, PassiveAggressive.Mode mode) Creates a new Passive Aggressive learner

Method Summary

Modifier and Type	Method and Description
CategoricalResults	classify(DataPoint data) Performs classification on the given data point.
SPA	clone()
double	getBias(int index) Returns the bias term used with the weight vector for the given class index.
double	getC() Returns the aggressiveness parameter
PassiveAggressive.Mode	getMode() Returns which version of the PA update is used
Vec	getRawWeight(int index) Returns the raw weight vector associated with the given class index.
static Distribution	guessC(DataSet d) Guess the distribution to use for the regularization term C in Support PassiveAggressive.
boolean	isUseBias() Returns true if an implicit bias term will be added, false otherwise
int	numWeightsVecs() Returns the number of weight vectors that can be returned.
void	setC(double C) Set the aggressiveness parameter.
void	setMode(PassiveAggressive.Mode mode) Sets which version of the PA update is used.
void	setUp(CategoricalData[] categoricalAttributes, int numericAttributes, CategoricalData predicting) Prepares the classifier to begin learning from its UpdateableClassifier.update(jsat.classifiers.DataPoint, int) method.
void	setUseBias(boolean useBias) Sets whether or not the implementation will use an implicit bias term appended to the inputs or not.
boolean	supportsWeightedData() Indicates whether the model knows how to train using weighted data points.
void	update(DataPoint dataPoint, int targetClass) Updates the classifier by giving it a new data point to learn from.

Methods inherited from class jsat.classifiers.BaseUpdateableClassifier

getEpochs, setEpochs, train, train, trainEpochs

Methods inherited from class java.lang.Object

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

Methods inherited from interface jsat.parameters.Parameterized

getParameter, getParameters

Constructor Detail

SPA

public SPA()

Creates a new Passive Aggressive learner that does 10 epochs and uses PA2.

SPA

public SPA(int epochs,
           PassiveAggressive.Mode mode)

Creates a new Passive Aggressive learner

Parameters:

epochs - the number of training epochs to use during batch training

mode - which version of the update to perform

Method Detail

setUseBias

public void setUseBias(boolean useBias)

Sets whether or not the implementation will use an implicit bias term appended to the inputs or not.

Parameters:

useBias - true to add an implicit bias term, false to use the data as given

isUseBias

public boolean isUseBias()

Returns true if an implicit bias term will be added, false otherwise

Returns:

true if an implicit bias term will be added, false otherwise

setC

public void setC(double C)

Set the aggressiveness parameter. Increasing the value of this parameter increases the aggressiveness of the algorithm. It must be a positive value. This parameter essentially performs a type of regularization on the updates
An infinitely large value is equivalent to being completely aggressive, and is performed when the mode is set to PassiveAggressive.Mode.PA.

Parameters:

C - the positive aggressiveness parameter

getC

public double getC()

Returns the aggressiveness parameter

Returns:

the aggressiveness parameter

setMode

public void setMode(PassiveAggressive.Mode mode)

Sets which version of the PA update is used.

Parameters:

mode - which PA update style to perform

getMode

public PassiveAggressive.Mode getMode()

Returns which version of the PA update is used

Returns:

which PA update style is used

getRawWeight

public Vec getRawWeight(int index)

Description copied from interface: SimpleWeightVectorModel

Returns the raw weight vector associated with the given class index. If the given class is an implicit zero vector, a ConstantVector object may be returned.
Do not alter the returned weight vector, as it will change the model's values.

If a regression problem, only index = 0 should be used

Specified by:

getRawWeight in interface SimpleWeightVectorModel

Parameters:

index - the class index to get the weight vector for

Returns:

the weight vector used for the specified class

getBias

public double getBias(int index)

Description copied from interface: SimpleWeightVectorModel

Returns the bias term used with the weight vector for the given class index. If the model does not support or was not trained with bias weights, 0 will be returned.

If a regression problem, only index = 0 should be used

Specified by:

getBias in interface SimpleWeightVectorModel

Parameters:

index - the class index to get the weight vector for

Returns:

the bias term for the specified class

numWeightsVecs

public int numWeightsVecs()

Description copied from interface: SimpleWeightVectorModel

Returns the number of weight vectors that can be returned. For binary classification problems the value may be 1 if only a single weight vector's sign is used to determine the class. For multi-class problems, the weight vector count includes the implicit zero vector (if one is being used).

Specified by:

numWeightsVecs in interface SimpleWeightVectorModel

Returns:

the number of weight vectors for which SimpleWeightVectorModel.getRawWeight(int) can be called.

clone

public SPA clone()

Specified by:

clone in interface Classifier

Specified by:

clone in interface UpdateableClassifier

Specified by:

clone in class BaseUpdateableClassifier

setUp

public void setUp(CategoricalData[] categoricalAttributes,
                  int numericAttributes,
                  CategoricalData predicting)

Description copied from interface: UpdateableClassifier

Prepares the classifier to begin learning from its UpdateableClassifier.update(jsat.classifiers.DataPoint, int) method.

Specified by:

setUp in interface UpdateableClassifier

Parameters:

categoricalAttributes - an array containing the categorical attributes that will be in each data point

numericAttributes - the number of numeric attributes that will be in each data point

predicting - the information for the target class that will be predicted

update

public void update(DataPoint dataPoint,
                   int targetClass)

Description copied from interface: UpdateableClassifier

Updates the classifier by giving it a new data point to learn from.

Specified by:

update in interface UpdateableClassifier

Parameters:

dataPoint - the data point to learn

targetClass - the target class of the data point

classify

public CategoricalResults classify(DataPoint data)

Description copied from interface: Classifier

Performs classification on the given data point.

Specified by:

classify in interface Classifier

Parameters:

data - the data point to classify

Returns:

the results of the classification.

supportsWeightedData

public boolean supportsWeightedData()

Description copied from interface: Classifier

Indicates whether the model knows how to train using weighted data points. If it does, the model will train assuming the weights. The values returned by this method may change depending on the parameters set for the model.

Specified by:

supportsWeightedData in interface Classifier

Returns:

true if the model supports weighted data, false otherwise

guessC

public static Distribution guessC(DataSet d)

Guess the distribution to use for the regularization term C in Support PassiveAggressive.

Parameters:

d - the data set to get the guess for

Returns:

the guess for the C parameter