gov.sandia.cognition.statistics.distribution

Class UnivariateGaussian.IncrementalEstimator

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.learning.algorithm.AbstractBatchAndIncrementalLearner<DataType,SufficientStatisticsType>

gov.sandia.cognition.statistics.AbstractIncrementalEstimator<java.lang.Double,UnivariateGaussian,UnivariateGaussian.SufficientStatistic>

gov.sandia.cognition.statistics.distribution.UnivariateGaussian.IncrementalEstimator

All Implemented Interfaces:

BatchAndIncrementalLearner<java.lang.Double,UnivariateGaussian.SufficientStatistic>, BatchLearner<java.util.Collection<? extends java.lang.Double>,UnivariateGaussian.SufficientStatistic>, IncrementalLearner<java.lang.Double,UnivariateGaussian.SufficientStatistic>, IncrementalEstimator<java.lang.Double,UnivariateGaussian,UnivariateGaussian.SufficientStatistic>, CloneableSerializable, java.io.Serializable, java.lang.Cloneable

Enclosing class:

UnivariateGaussian

public static class UnivariateGaussian.IncrementalEstimator
extends AbstractIncrementalEstimator<java.lang.Double,UnivariateGaussian,UnivariateGaussian.SufficientStatistic>

Implements an incremental estimator for the sufficient statistics for a UnivariateGaussian.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static double	DEFAULT_DEFAULT_VARIANCE The default value for the default variance is 1.0E-5.
protected double	defaultVariance Amount to add to the variance to keep it from being 0.0.

Constructor Summary

Constructors

Constructor and Description
IncrementalEstimator() Creates a new IncrementalEstimator.
IncrementalEstimator(double defaultVariance) Creates a new IncrementalEstimator with the given default variance.

Method Summary

Modifier and Type	Method and Description
UnivariateGaussian.SufficientStatistic	createInitialLearnedObject() Creates a new initial learned object, before any data is given.
double	getDefaultVariance() Gets the default variance, which is the amount added to the variance.
void	setDefaultVariance(double defaultVariance) Sets the default variance, which is the amount added to the variance.

Methods inherited from class gov.sandia.cognition.statistics.AbstractIncrementalEstimator

clone, update

Methods inherited from class gov.sandia.cognition.learning.algorithm.AbstractBatchAndIncrementalLearner

learn, learn, update

Methods inherited from class java.lang.Object

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

Methods inherited from interface gov.sandia.cognition.learning.algorithm.BatchAndIncrementalLearner

learn

Methods inherited from interface gov.sandia.cognition.learning.algorithm.BatchLearner

learn

Methods inherited from interface gov.sandia.cognition.learning.algorithm.IncrementalLearner

update

Field Detail

DEFAULT_DEFAULT_VARIANCE

public static final double DEFAULT_DEFAULT_VARIANCE

The default value for the default variance is 1.0E-5.

See Also:

Constant Field Values

defaultVariance

protected double defaultVariance

Amount to add to the variance to keep it from being 0.0.

Constructor Detail

IncrementalEstimator

public IncrementalEstimator()

Creates a new IncrementalEstimator.

IncrementalEstimator

public IncrementalEstimator(double defaultVariance)

Creates a new IncrementalEstimator with the given default variance.

Parameters:

defaultVariance - The default variance. Cannot be negative.

Method Detail

createInitialLearnedObject

public UnivariateGaussian.SufficientStatistic createInitialLearnedObject()

Description copied from interface: IncrementalLearner

Creates a new initial learned object, before any data is given.

Returns:

The initial learned object.

getDefaultVariance

public double getDefaultVariance()

Gets the default variance, which is the amount added to the variance. If this is greater than zero, it avoids creating zero variance.

Returns:

The default variance. Cannot be negative.

setDefaultVariance

public void setDefaultVariance(double defaultVariance)

Sets the default variance, which is the amount added to the variance. If this is greater than zero, it avoids creating zero variance.

Parameters:

defaultVariance - The default variance. Cannot be negative.