gov.sandia.cognition.learning.function.categorization
Interface ConfidenceWeightedBinaryCategorizer

All Superinterfaces:
BinaryCategorizer<Vectorizable>, Categorizer<Vectorizable,Boolean>, Cloneable, CloneableSerializable, DiscriminantBinaryCategorizer<Vectorizable>, DiscriminantCategorizer<Vectorizable,Boolean,Double>, Evaluator<Vectorizable,Boolean>, Serializable, ThresholdBinaryCategorizer<Vectorizable>, VectorInputEvaluator<Vectorizable,Boolean>
All Known Implementing Classes:
AbstractConfidenceWeightedBinaryCategorizer, DefaultConfidenceWeightedBinaryCategorizer, DiagonalConfidenceWeightedBinaryCategorizer

public interface ConfidenceWeightedBinaryCategorizer
extends VectorInputEvaluator<Vectorizable,Boolean>, ThresholdBinaryCategorizer<Vectorizable>

Interface for a confidence-weighted binary categorizer, which defines a distribution over linear binary categorizers. It extends the vector input evaluator and threshold binary categorizer like a LinearBinaryCategorizer so that it can behave as a binary categorizer, it but also has methods for accessing the distribution of binary categorizers that it represents. It is typically represented using a mean vector and a covariance matrix.

Since:
3.3.0
Author:
Justin Basilico

Method Summary
 MultivariateGaussian createWeightDistribution()
          Creates a multivariate Gaussian distribution that represents the distribution of weight vectors that the algorithm has learned.
 BernoulliDistribution evaluateAsBernoulli(Vectorizable input)
          Returns a Bernoulli distribution over the output of the distribution of weight vectors times the input, with the confidence that the categorizer was trained using.
 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.
 Vector getMean()
          Gets the mean of the categorizer, which is the weight vector.
 boolean isInitialized()
          Determines if this category has been initialized with a mean and covariance.
 
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
 

Method Detail

evaluateAsGaussian

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.

Parameters:
input - The input to evaluate.
Returns:
The distribution of outputs as a Gaussian.

evaluateAsBernoulli

BernoulliDistribution evaluateAsBernoulli(Vectorizable input)
Returns a Bernoulli 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 over outputs as a Bernoulli.

createWeightDistribution

MultivariateGaussian createWeightDistribution()
Creates a multivariate Gaussian distribution that represents the distribution of weight vectors that the algorithm has learned.

Returns:
The distribution of weight vectors.

isInitialized

boolean isInitialized()
Determines if this category has been initialized with a mean and covariance.

Returns:
True if this categorizer has been initialized. Otherwise, false.

getMean

Vector getMean()
Gets the mean of the categorizer, which is the weight vector.

Returns:
The mean of the categorizer.

getCovariance

Matrix getCovariance()
Gets the covariance matrix of the categorizer.

Returns:
The covariance matrix.