weka.classifiers.pmml.consumer

Class PMMLClassifier

java.lang.Object

weka.classifiers.Classifier

weka.classifiers.pmml.consumer.PMMLClassifier

All Implemented Interfaces:

Serializable, Cloneable, CapabilitiesHandler, OptionHandler, PMMLModel, RevisionHandler

Direct Known Subclasses:

GeneralRegression, NeuralNetwork, Regression

public abstract class PMMLClassifier
extends Classifier
implements Serializable, PMMLModel

Abstract base class for all PMML classifiers.

Version:

$Revision: 5562 $

Author:

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

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected String	m_creatorApplication Creator application
protected Instances	m_dataDictionary The data dictionary
protected MappingInfo	m_fieldsMap The mapping between mining schema fields and incoming instance attributes
protected boolean	m_initialized Has the classifier been initialized (i.e.
protected Logger	m_log Logger
protected MiningSchema	m_miningSchema The fields and meta data used by the model
protected String	m_pmmlVersion PMML version

Fields inherited from class weka.classifiers.Classifier

m_Debug

Method Summary

Modifier and Type	Method and Description
void	buildClassifier(Instances data) Throw an exception - PMML models are pre-built.
void	done() Signal that a scoring run has been completed.
String	getCreatorApplication() Get the name of the application that created this model
Instances	getDataDictionary() Get the data dictionary.
String	getFieldsMappingString() Get a textual description of the mapping between mining schema fields and incoming data fields.
Logger	getLog() Get the logger.
MiningSchema	getMiningSchema() Get the mining schema for this model.
String	getPMMLVersion() Get the PMML version used for this model.
void	mapToMiningSchema(Instances dataSet) Map mining schema to incoming instances.
void	setCreatorApplication(Document doc) Set the name of the application (if specified) that created this model
void	setLog(Logger log) Set a logger to use.
void	setPMMLVersion(Document doc) Set the version of PMML used for this model.

Methods inherited from class weka.classifiers.Classifier

classifyInstance, debugTipText, distributionForInstance, forName, getCapabilities, getDebug, getOptions, getRevision, listOptions, makeCopies, makeCopy, runClassifier, setDebug, setOptions

Methods inherited from class java.lang.Object

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

Field Detail

m_pmmlVersion

protected String m_pmmlVersion

PMML version

m_creatorApplication

protected String m_creatorApplication

Creator application

m_log

protected Logger m_log

Logger

m_dataDictionary

protected Instances m_dataDictionary

The data dictionary

m_miningSchema

protected MiningSchema m_miningSchema

The fields and meta data used by the model

m_fieldsMap

protected transient MappingInfo m_fieldsMap

The mapping between mining schema fields and incoming instance attributes

m_initialized

protected transient boolean m_initialized

Has the classifier been initialized (i.e. have we established a mapping between the mining schema and the incoming instances)?

Method Detail

setPMMLVersion

public void setPMMLVersion(Document doc)

Set the version of PMML used for this model.

Specified by:

setPMMLVersion in interface PMMLModel

Parameters:

doc - the Document encapsulating the pmml

setCreatorApplication

public void setCreatorApplication(Document doc)

Set the name of the application (if specified) that created this model

Specified by:

setCreatorApplication in interface PMMLModel

Parameters:

doc - the Document encapsulating the pmml

getDataDictionary

public Instances getDataDictionary()

Get the data dictionary.

Returns:

the data dictionary

getMiningSchema

public MiningSchema getMiningSchema()

Get the mining schema for this model.

Specified by:

getMiningSchema in interface PMMLModel

Returns:

the mining schema

getPMMLVersion

public String getPMMLVersion()

Get the PMML version used for this model.

Specified by:

getPMMLVersion in interface PMMLModel

Returns:

the PMML version

getCreatorApplication

public String getCreatorApplication()

Get the name of the application that created this model

Specified by:

getCreatorApplication in interface PMMLModel

Returns:

the name of the creating application or null if not specified in the pmml.

setLog

public void setLog(Logger log)

Set a logger to use.

Specified by:

setLog in interface PMMLModel

Parameters:

log - the logger to use

getLog

public Logger getLog()

Get the logger.

Specified by:

getLog in interface PMMLModel

Returns:

the logger (or null if none is being used)

buildClassifier

public void buildClassifier(Instances data)
                     throws Exception

Throw an exception - PMML models are pre-built.

Specified by:

buildClassifier in class Classifier

Parameters:

data - the Instances to learn from

Throws:

Exception - if something goes wrong

done

public void done()

Signal that a scoring run has been completed. Resets the initialized state to false so that a subsequent scoring run will trigger the mapping of the mining schema to incoming instances. If not called after a scoring run, then the classifier will assume that the current mapping is still valid.

mapToMiningSchema

public void mapToMiningSchema(Instances dataSet)
                       throws Exception

Map mining schema to incoming instances.

Parameters:

dataSet - the structure of the incoming Instances

Throws:

Exception - if something goes wrong

getFieldsMappingString

public String getFieldsMappingString()

Get a textual description of the mapping between mining schema fields and incoming data fields.

Returns:

a description of the fields mapping as a String or null if no mapping has been constructed yet.