Class DiscreteEstimatorBayes
java.lang.Object
weka.estimators.Estimator
weka.classifiers.bayes.net.estimate.DiscreteEstimatorBayes
- All Implemented Interfaces:
Serializable,Cloneable,Scoreable,CapabilitiesHandler,CapabilitiesIgnorer,OptionHandler,RevisionHandler
- Direct Known Subclasses:
DiscreteEstimatorFullBayes
Symbolic probability estimator based on symbol counts and a prior.
- Version:
- $Revision: 8034 $
- Author:
- Remco Bouckaert (rrb@xm.co.nz)
- See Also:
-
Field Summary
-
Constructor Summary
Constructors -
Method Summary
Modifier and TypeMethodDescriptionvoidaddValue(double data, double weight) Add a new data value to the current estimator.doublegetCount(double data) Get a counts for a valueintGets the number of symbols this estimator operates withdoublegetProbability(double data) Get a probability estimate for a valueReturns the revision string.doublelogScore(int nType, int nCardinality) Gets the log score contribution of this distributionstatic voidMain method for testing this class.toString()Display a representation of this estimatorMethods inherited from class weka.estimators.Estimator
addValues, addValues, addValues, addValues, buildEstimator, buildEstimator, clone, debugTipText, doNotCheckCapabilitiesTipText, equals, forName, getCapabilities, getDebug, getDoNotCheckCapabilities, getOptions, listOptions, makeCopies, makeCopy, setDebug, setDoNotCheckCapabilities, setOptions, testCapabilities
-
Constructor Details
-
DiscreteEstimatorBayes
public DiscreteEstimatorBayes(int nSymbols, double fPrior) Constructor- Parameters:
nSymbols- the number of possible symbols (remember to include 0)fPrior-
-
-
Method Details
-
addValue
public void addValue(double data, double weight) Add a new data value to the current estimator. -
getProbability
public double getProbability(double data) Get a probability estimate for a value- Specified by:
getProbabilityin classEstimator- Parameters:
data- the value to estimate the probability of- Returns:
- the estimated probability of the supplied value
-
getCount
public double getCount(double data) Get a counts for a value- Parameters:
data- the value to get the counts for- Returns:
- the count of the supplied value
-
getNumSymbols
public int getNumSymbols()Gets the number of symbols this estimator operates with- Returns:
- the number of estimator symbols
-
logScore
public double logScore(int nType, int nCardinality) Gets the log score contribution of this distribution -
toString
Display a representation of this estimator -
getRevision
Returns the revision string.- Specified by:
getRevisionin interfaceRevisionHandler- Overrides:
getRevisionin classEstimator- Returns:
- the revision
-
main
Main method for testing this class.- Parameters:
argv- should contain a sequence of integers which will be treated as symbolic.
-