All Packages  Class Hierarchy  This Package  Previous  Next  Index  WEKA's home

Class weka.classifiers.HyperPipes

java.lang.Object
   |
   +----weka.classifiers.Classifier
           |
           +----weka.classifiers.DistributionClassifier
                   |
                   +----weka.classifiers.HyperPipes

public class HyperPipes
extends DistributionClassifier
Class implementing a HyperPipe classifier. For each category a HyperPipe is constructed that contains all points of that category (essentially records the attribute bounds observed for each category). Test instances are classified according to the category that most contains the instance). Does not handle numeric class, or missing values in test cases. Extremely simple algorithm, but has the advantage of being extremely fast, and works quite well when you have smegloads of attributes.

Author:
Lucio de Souza Coelho (lucio@intelligenesis.net), Len Trigg (len@intelligenesis.net)

Constructor Index

 o HyperPipes()

Method Index

 o buildClassifier(Instances)
Generates the classifier.
 o distributionForInstance(Instance)
Classifies the given test instance.
 o main(String[])
Main method for testing this class.
 o toString()
Returns a description of this classifier.
 o updateClassifier(Instance)
Updates the classifier.

Constructors

 o HyperPipes
 public HyperPipes()

Methods

 o buildClassifier
 public void buildClassifier(Instances instances) throws Exception
Generates the classifier.

Parameters:
instances - set of instances serving as training data
Throws: Exception
if the classifier has not been generated successfully
Overrides:
buildClassifier in class Classifier
 o updateClassifier
 public void updateClassifier(Instance instance) throws Exception
Updates the classifier.

Parameters:
instance - the instance to be put into the classifier
Throws: Exception
if the instance could not be included successfully
 o distributionForInstance
 public double[] distributionForInstance(Instance instance) throws Exception
Classifies the given test instance.

Parameters:
instance - the instance to be classified
Returns:
the predicted class for the instance
Throws: Exception
if the instance can't be classified
Overrides:
distributionForInstance in class DistributionClassifier
 o toString
 public String toString()
Returns a description of this classifier.

Returns:
a description of this classifier as a string.
Overrides:
toString in class Object
 o main
 public static void main(String argv[])
Main method for testing this class.

Parameters:
argv - should contain command line arguments for evaluation (see Evaluation).

All Packages  Class Hierarchy  This Package  Previous  Next  Index  WEKA's home