gov.sandia.cognition.statistics.distribution
Class MultivariateGaussian.MaximumLikelihoodEstimator

java.lang.Object
  extended by gov.sandia.cognition.util.AbstractCloneableSerializable
      extended by gov.sandia.cognition.statistics.distribution.MultivariateGaussian.MaximumLikelihoodEstimator
All Implemented Interfaces:
BatchLearner<Collection<? extends Vector>,MultivariateGaussian.PDF>, DistributionEstimator<Vector,MultivariateGaussian.PDF>, CloneableSerializable, Serializable, Cloneable
Enclosing class:
MultivariateGaussian

public static class MultivariateGaussian.MaximumLikelihoodEstimator
extends AbstractCloneableSerializable
implements DistributionEstimator<Vector,MultivariateGaussian.PDF>

Computes the Maximum Likelihood Estimate of the MultivariateGaussian given a set of Vectors

See Also:
Serialized Form

Field Summary
static double DEFAULT_COVARIANCE
          Default covariance used in estimation, 1.0E-5.
 
Constructor Summary
MultivariateGaussian.MaximumLikelihoodEstimator()
          Default constructor;
MultivariateGaussian.MaximumLikelihoodEstimator(double defaultCovariance)
          Creates a new instance of MaximumLikelihoodEstimator
 
Method Summary
 MultivariateGaussian.PDF learn(Collection<? extends Vector> data)
          Computes the Gaussian that estimates the maximum likelihood of generating the given set of samples.
static MultivariateGaussian.PDF learn(Collection<? extends Vector> data, double defaultCovariance)
          Computes the Gaussian that estimates the maximum likelihood of generating the given set of samples.
 
Methods inherited from class gov.sandia.cognition.util.AbstractCloneableSerializable
clone
 
Methods inherited from class java.lang.Object
equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 
Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable
clone
 

Field Detail

DEFAULT_COVARIANCE

public static final double DEFAULT_COVARIANCE
Default covariance used in estimation, 1.0E-5.

See Also:
Constant Field Values
Constructor Detail

MultivariateGaussian.MaximumLikelihoodEstimator

public MultivariateGaussian.MaximumLikelihoodEstimator()
Default constructor;


MultivariateGaussian.MaximumLikelihoodEstimator

public MultivariateGaussian.MaximumLikelihoodEstimator(double defaultCovariance)
Creates a new instance of MaximumLikelihoodEstimator

Parameters:
defaultCovariance - Amount to add to the diagonal of the covariance matrix
Method Detail

learn

public static MultivariateGaussian.PDF learn(Collection<? extends Vector> data,
                                             double defaultCovariance)
Computes the Gaussian that estimates the maximum likelihood of generating the given set of samples.

Parameters:
defaultCovariance - amount to add to the diagonals of the covariance matrix, typically on the order of 1e-4, can be 0.0.
data - The samples to calculate the Gaussian from throws IllegalArgumentException if samples has 1 or fewer samples.
Returns:
The Gaussian that estimates the maximum likelihood of generating the given data.

learn

public MultivariateGaussian.PDF learn(Collection<? extends Vector> data)
Computes the Gaussian that estimates the maximum likelihood of generating the given set of samples.

Specified by:
learn in interface BatchLearner<Collection<? extends Vector>,MultivariateGaussian.PDF>
Parameters:
data - The samples to calculate the Gaussian from throws IllegalArgumentException if samples has 1 or fewer samples.
Returns:
The Gaussian that estimates the maximum likelihood of generating the given data.