weka.distributed.clusterers

Class CanopyBuilder

java.lang.Object

weka.distributed.clusterers.CanopyBuilder

All Implemented Interfaces:

java.io.Serializable

public class CanopyBuilder
extends java.lang.Object
implements java.io.Serializable

Helper class for building a distributed canopy clusterer

Version:

$Revision: 11631 $

Author:

Mark Hall (mhall{[at]}pentaho{[dot]}com)

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
CanopyBuilder(weka.core.Instances headerWithSummary, weka.core.Instances headerNoSummary, CanopyMapTask canopyMapTask, java.lang.String csvMapTaskOptions) Construct a new Canopy builder

Method Summary

Modifier and Type	Method and Description
void	finishedInput() Notify that there is no more input
weka.clusterers.Clusterer	getFinalizedClusterer() Return the finalized clusterer.
void	process(java.lang.Object current) Process the current object (instance or string).

Methods inherited from class java.lang.Object

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

Constructor Detail

CanopyBuilder

public CanopyBuilder(weka.core.Instances headerWithSummary,
                     weka.core.Instances headerNoSummary,
                     CanopyMapTask canopyMapTask,
                     java.lang.String csvMapTaskOptions)

Construct a new Canopy builder

Parameters:

headerWithSummary - the header of the data with summary attributes

headerNoSummary - the header of the data without summary attributes

canopyMapTask - the configured map task to use

csvMapTaskOptions - options for csv parsing

Method Detail

process

public void process(java.lang.Object current)
             throws DistributedWekaException,
                    java.io.IOException

Process the current object (instance or string).

Parameters:

current - either an Instance or a string (assumed CSV)

Throws:

DistributedWekaException - if a problem occurs

java.io.IOException - if a problem occurs

finishedInput

public void finishedInput()

Notify that there is no more input

getFinalizedClusterer

public weka.clusterers.Clusterer getFinalizedClusterer()
                                                throws DistributedWekaException

Return the finalized clusterer. This method should only be called after finishedInput().

Returns:

the finalized clusterer

Throws:

DistributedWekaException - if a problem occurs