gov.sandia.cognition.learning.function.categorization

Class DefaultConfidenceWeightedBinaryCategorizer

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.learning.function.categorization.AbstractBinaryCategorizer<InputType>

gov.sandia.cognition.learning.function.categorization.AbstractDiscriminantBinaryCategorizer<Vectorizable>

gov.sandia.cognition.learning.function.categorization.LinearBinaryCategorizer

gov.sandia.cognition.learning.function.categorization.AbstractConfidenceWeightedBinaryCategorizer

gov.sandia.cognition.learning.function.categorization.DefaultConfidenceWeightedBinaryCategorizer

All Implemented Interfaces:

Evaluator<Vectorizable,java.lang.Boolean>, BinaryCategorizer<Vectorizable>, Categorizer<Vectorizable,java.lang.Boolean>, ConfidenceWeightedBinaryCategorizer, DiscriminantBinaryCategorizer<Vectorizable>, DiscriminantCategorizer<Vectorizable,java.lang.Boolean,java.lang.Double>, ThresholdBinaryCategorizer<Vectorizable>, VectorInputEvaluator<Vectorizable,java.lang.Boolean>, CloneableSerializable, java.io.Serializable, java.lang.Cloneable

public class DefaultConfidenceWeightedBinaryCategorizer
extends AbstractConfidenceWeightedBinaryCategorizer

A default implementation of the ConfidenceWeightedBinaryCategorizer that stores a full mean and covariance matrix.

Since:

3.3.0

Author:

Justin Basilico

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected Matrix	covariance The covariance matrix.

Fields inherited from class gov.sandia.cognition.learning.function.categorization.LinearBinaryCategorizer

bias, DEFAULT_BIAS, weights

Fields inherited from class gov.sandia.cognition.learning.function.categorization.AbstractBinaryCategorizer

BINARY_CATEGORIES

Constructor Summary

Constructors

Constructor and Description
DefaultConfidenceWeightedBinaryCategorizer() Creates a new, uninitialized DefaultConfidenceWeightedBinaryCategorizer.
DefaultConfidenceWeightedBinaryCategorizer(Vector mean, Matrix covariance) Creates a new DefaultConfidenceWeightedBinaryCategorizer with the given mean and covariance.

Method Summary

Modifier and Type	Method and Description
UnivariateGaussian	evaluateAsGaussian(Vectorizable input) Returns the univariate Gaussian distribution over the output of the distribution of weight vectors times the input, with the confidence that the categorizer was trained using.
Matrix	getCovariance() Gets the covariance matrix of the categorizer.
void	setCovariance(Matrix covariance) Sets the covariance matrix.

Methods inherited from class gov.sandia.cognition.learning.function.categorization.AbstractConfidenceWeightedBinaryCategorizer

createWeightDistribution, evaluateAsBernoulli, getMean, isInitialized, setMean

Methods inherited from class gov.sandia.cognition.learning.function.categorization.LinearBinaryCategorizer

clone, evaluateAsDouble, evaluateAsDouble, getBias, getInputDimensionality, getThreshold, getWeights, setBias, setThreshold, setWeights, toString

Methods inherited from class gov.sandia.cognition.learning.function.categorization.AbstractDiscriminantBinaryCategorizer

evaluate, evaluateWithDiscriminant

Methods inherited from class gov.sandia.cognition.learning.function.categorization.AbstractBinaryCategorizer

getCategories

Methods inherited from class java.lang.Object

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

Methods inherited from interface gov.sandia.cognition.math.matrix.VectorInputEvaluator

getInputDimensionality

Methods inherited from interface gov.sandia.cognition.learning.function.categorization.ThresholdBinaryCategorizer

getThreshold, setThreshold

Methods inherited from interface gov.sandia.cognition.learning.function.categorization.DiscriminantBinaryCategorizer

evaluateAsDouble

Methods inherited from interface gov.sandia.cognition.learning.function.categorization.DiscriminantCategorizer

evaluateWithDiscriminant

Methods inherited from interface gov.sandia.cognition.learning.function.categorization.Categorizer

getCategories

Methods inherited from interface gov.sandia.cognition.evaluator.Evaluator

evaluate

Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable

clone

Field Detail

covariance

protected Matrix covariance

The covariance matrix.

Constructor Detail

DefaultConfidenceWeightedBinaryCategorizer

public DefaultConfidenceWeightedBinaryCategorizer()

Creates a new, uninitialized DefaultConfidenceWeightedBinaryCategorizer.

DefaultConfidenceWeightedBinaryCategorizer

public DefaultConfidenceWeightedBinaryCategorizer(Vector mean,
                                                  Matrix covariance)

Creates a new DefaultConfidenceWeightedBinaryCategorizer with the given mean and covariance. The covariance matrix must be an d by d matrix where d is the dimensionality of the mean.

Parameters:

mean - The mean vector.

covariance - The covariance matrix.

Method Detail

evaluateAsGaussian

public UnivariateGaussian evaluateAsGaussian(Vectorizable input)

Description copied from interface: ConfidenceWeightedBinaryCategorizer

Returns the univariate Gaussian distribution over the output of the distribution of weight vectors times the input, with the confidence that the categorizer was trained using.

Parameters:

input - The input to evaluate.

Returns:

The distribution of outputs as a Gaussian.

getCovariance

public Matrix getCovariance()

Description copied from interface: ConfidenceWeightedBinaryCategorizer

Gets the covariance matrix of the categorizer.

Returns:

The covariance matrix.

setCovariance

public void setCovariance(Matrix covariance)

Sets the covariance matrix. Must be a square matrix the same size as the dimensionality of the mean.

Parameters:

covariance - The covariance matrix.