weka.experiment
Class CrossValidationResultProducer

java.lang.Object
  weka.experiment.CrossValidationResultProducer

All Implemented Interfaces:

Serializable, AdditionalMeasureProducer, OptionHandler, RevisionHandler, ResultProducer

Direct Known Subclasses:

CrossValidationSplitResultProducer

public class CrossValidationResultProducer
extends Object
implements ResultProducer, OptionHandler, AdditionalMeasureProducer, RevisionHandler

Generates for each run, carries out an n-fold cross-validation, using the set SplitEvaluator to generate some results. If the class attribute is nominal, the dataset is stratified. Results for each fold are generated, so you may wish to use this in addition with an AveragingResultProducer to obtain averages for each run.

Valid options are:

 -X <number of folds>
  The number of folds to use for the cross-validation.
  (default 10)

 -D
 Save raw split evaluator output.

 -O <file/directory name/path>
  The filename where raw output will be stored.
  If a directory name is specified then then individual
  outputs will be gzipped, otherwise all output will be
  zipped to the named file. Use in conjuction with -D. (default splitEvalutorOut.zip)

 -W <class name>
  The full class name of a SplitEvaluator.
  eg: weka.experiment.ClassifierSplitEvaluator

 
 Options specific to split evaluator weka.experiment.ClassifierSplitEvaluator:
 

 -W <class name>
  The full class name of the classifier.
  eg: weka.classifiers.bayes.NaiveBayes

 -C <index>
  The index of the class for which IR statistics
  are to be output. (default 1)

 -I <index>
  The index of an attribute to output in the
  results. This attribute should identify an
  instance in order to know which instances are
  in the test set of a cross validation. if 0
  no output (default 0).

 -P
  Add target and prediction columns to the result
  for each fold.

 
 Options specific to classifier weka.classifiers.rules.ZeroR:
 

 -D
  If set, classifier is run in debug mode and
  may output additional info to the console

All options after -- will be passed to the split evaluator.

Version:

$Revision: 8034 $

Author:

Len Trigg (trigg@cs.waikato.ac.nz)

See Also:

Serialized Form

Field Summary

static String	DATASET_FIELD_NAME The name of the key field containing the dataset name
static String	FOLD_FIELD_NAME The name of the key field containing the fold number
static String	RUN_FIELD_NAME The name of the key field containing the run number
static String	TIMESTAMP_FIELD_NAME The name of the result field containing the timestamp

Constructor Summary

CrossValidationResultProducer()

Method Summary

void	doRun(int run) Gets the results for a specified run number.
void	doRunKeys(int run) Gets the keys for a specified run number.
Enumeration	enumerateMeasures() Returns an enumeration of any additional measure names that might be in the SplitEvaluator
String	getCompatibilityState() Gets a description of the internal settings of the result producer, sufficient for distinguishing a ResultProducer instance from another with different settings (ignoring those settings set through this interface).
String[]	getKeyNames() Gets the names of each of the columns produced for a single run.
Object[]	getKeyTypes() Gets the data types of each of the columns produced for a single run.
double	getMeasure(String additionalMeasureName) Returns the value of the named measure
int	getNumFolds() Get the value of NumFolds.
String[]	getOptions() Gets the current settings of the result producer.
File	getOutputFile() Get the value of OutputFile.
boolean	getRawOutput() Get if raw split evaluator output is to be saved
String[]	getResultNames() Gets the names of each of the columns produced for a single run.
Object[]	getResultTypes() Gets the data types of each of the columns produced for a single run.
String	getRevision() Returns the revision string.
SplitEvaluator	getSplitEvaluator() Get the SplitEvaluator.
static Double	getTimestamp() Gets a Double representing the current date and time.
String	globalInfo() Returns a string describing this result producer
Enumeration	listOptions() Returns an enumeration describing the available options..
static void	main(String[] args) Quick test of timestamp
String	numFoldsTipText() Returns the tip text for this property
String	outputFileTipText() Returns the tip text for this property
void	postProcess() Perform any postprocessing.
void	preProcess() Prepare to generate results.
String	rawOutputTipText() Returns the tip text for this property
void	setAdditionalMeasures(String[] additionalMeasures) Set a list of method names for additional measures to look for in SplitEvaluators.
void	setInstances(Instances instances) Sets the dataset that results will be obtained for.
void	setNumFolds(int newNumFolds) Set the value of NumFolds.
void	setOptions(String[] options) Parses a given list of options.
void	setOutputFile(File newOutputFile) Set the value of OutputFile.
void	setRawOutput(boolean d) Set to true if raw split evaluator output is to be saved
void	setResultListener(ResultListener listener) Sets the object to send results of each run to.
void	setSplitEvaluator(SplitEvaluator newSplitEvaluator) Set the SplitEvaluator.
String	splitEvaluatorTipText() Returns the tip text for this property
String	toString() Gets a text descrption of the result producer.

Methods inherited from class java.lang.Object

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

Field Detail

DATASET_FIELD_NAME

public static String DATASET_FIELD_NAME

The name of the key field containing the dataset name

RUN_FIELD_NAME

public static String RUN_FIELD_NAME

The name of the key field containing the run number

FOLD_FIELD_NAME

public static String FOLD_FIELD_NAME

The name of the key field containing the fold number

TIMESTAMP_FIELD_NAME

public static String TIMESTAMP_FIELD_NAME

The name of the result field containing the timestamp

Constructor Detail

CrossValidationResultProducer

public CrossValidationResultProducer()

Method Detail

globalInfo

public String globalInfo()

Returns a string describing this result producer

Returns:

a description of the result producer suitable for displaying in the explorer/experimenter gui

setInstances

public void setInstances(Instances instances)

Sets the dataset that results will be obtained for.

Specified by:

setInstances in interface ResultProducer

Parameters:

instances - a value of type 'Instances'.

setResultListener

public void setResultListener(ResultListener listener)

Sets the object to send results of each run to.

Specified by:

setResultListener in interface ResultProducer

Parameters:

listener - a value of type 'ResultListener'

setAdditionalMeasures

public void setAdditionalMeasures(String[] additionalMeasures)

Set a list of method names for additional measures to look for in SplitEvaluators. This could contain many measures (of which only a subset may be produceable by the current SplitEvaluator) if an experiment is the type that iterates over a set of properties.

Specified by:

setAdditionalMeasures in interface ResultProducer

Parameters:

additionalMeasures - an array of measure names, null if none

enumerateMeasures

public Enumeration enumerateMeasures()

Returns an enumeration of any additional measure names that might be in the SplitEvaluator

Specified by:

enumerateMeasures in interface AdditionalMeasureProducer

Returns:

an enumeration of the measure names

getMeasure

public double getMeasure(String additionalMeasureName)

Returns the value of the named measure

Specified by:

getMeasure in interface AdditionalMeasureProducer

Parameters:

additionalMeasureName - the name of the measure to query for its value

Returns:

the value of the named measure

Throws:

IllegalArgumentException - if the named measure is not supported

getTimestamp

public static Double getTimestamp()

Gets a Double representing the current date and time. eg: 1:46pm on 20/5/1999 -> 19990520.1346

Returns:

a value of type Double

preProcess

public void preProcess()
                throws Exception

Prepare to generate results.

Specified by:

preProcess in interface ResultProducer

Throws:

Exception - if an error occurs during preprocessing.

postProcess

public void postProcess()
                 throws Exception

Perform any postprocessing. When this method is called, it indicates that no more requests to generate results for the current experiment will be sent.

Specified by:

postProcess in interface ResultProducer

Throws:

Exception - if an error occurs

doRunKeys

public void doRunKeys(int run)
               throws Exception

Gets the keys for a specified run number. Different run numbers correspond to different randomizations of the data. Keys produced should be sent to the current ResultListener

Specified by:

doRunKeys in interface ResultProducer

Parameters:

run - the run number to get keys for.

Throws:

Exception - if a problem occurs while getting the keys

doRun

public void doRun(int run)
           throws Exception

Gets the results for a specified run number. Different run numbers correspond to different randomizations of the data. Results produced should be sent to the current ResultListener

Specified by:

doRun in interface ResultProducer

Parameters:

run - the run number to get results for.

Throws:

Exception - if a problem occurs while getting the results

getKeyNames

public String[] getKeyNames()

Gets the names of each of the columns produced for a single run. This method should really be static.

Specified by:

getKeyNames in interface ResultProducer

Returns:

an array containing the name of each column

getKeyTypes

public Object[] getKeyTypes()

Gets the data types of each of the columns produced for a single run. This method should really be static.

Specified by:

getKeyTypes in interface ResultProducer

Returns:

an array containing objects of the type of each column. The objects should be Strings, or Doubles.

getResultNames

public String[] getResultNames()

Gets the names of each of the columns produced for a single run. This method should really be static.

Specified by:

getResultNames in interface ResultProducer

Returns:

an array containing the name of each column

getResultTypes

public Object[] getResultTypes()

Gets the data types of each of the columns produced for a single run. This method should really be static.

Specified by:

getResultTypes in interface ResultProducer

Returns:

an array containing objects of the type of each column. The objects should be Strings, or Doubles.

getCompatibilityState

public String getCompatibilityState()

Gets a description of the internal settings of the result producer, sufficient for distinguishing a ResultProducer instance from another with different settings (ignoring those settings set through this interface). For example, a cross-validation ResultProducer may have a setting for the number of folds. For a given state, the results produced should be compatible. Typically if a ResultProducer is an OptionHandler, this string will represent the command line arguments required to set the ResultProducer to that state.

Specified by:

getCompatibilityState in interface ResultProducer

Returns:

the description of the ResultProducer state, or null if no state is defined

outputFileTipText

public String outputFileTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getOutputFile

public File getOutputFile()

Get the value of OutputFile.

Returns:

Value of OutputFile.

setOutputFile

public void setOutputFile(File newOutputFile)

Set the value of OutputFile.

Parameters:

newOutputFile - Value to assign to OutputFile.

numFoldsTipText

public String numFoldsTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getNumFolds

public int getNumFolds()

Get the value of NumFolds.

Returns:

Value of NumFolds.

setNumFolds

public void setNumFolds(int newNumFolds)

Set the value of NumFolds.

Parameters:

newNumFolds - Value to assign to NumFolds.

rawOutputTipText

public String rawOutputTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getRawOutput

public boolean getRawOutput()

Get if raw split evaluator output is to be saved

Returns:

true if raw split evalutor output is to be saved

setRawOutput

public void setRawOutput(boolean d)

Set to true if raw split evaluator output is to be saved

Parameters:

d - true if output is to be saved

splitEvaluatorTipText

public String splitEvaluatorTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getSplitEvaluator

public SplitEvaluator getSplitEvaluator()

Get the SplitEvaluator.

Returns:

the SplitEvaluator.

setSplitEvaluator

public void setSplitEvaluator(SplitEvaluator newSplitEvaluator)

Set the SplitEvaluator.

Parameters:

newSplitEvaluator - new SplitEvaluator to use.

listOptions

public Enumeration listOptions()

Returns an enumeration describing the available options..

Specified by:

listOptions in interface OptionHandler

Returns:

an enumeration of all the available options.

setOptions

public void setOptions(String[] options)
                throws Exception

Parses a given list of options.

Valid options are:

 -X <number of folds>
  The number of folds to use for the cross-validation.
  (default 10)

 -D
 Save raw split evaluator output.

 -O <file/directory name/path>
  The filename where raw output will be stored.
  If a directory name is specified then then individual
  outputs will be gzipped, otherwise all output will be
  zipped to the named file. Use in conjuction with -D. (default splitEvalutorOut.zip)

 -W <class name>
  The full class name of a SplitEvaluator.
  eg: weka.experiment.ClassifierSplitEvaluator

 
 Options specific to split evaluator weka.experiment.ClassifierSplitEvaluator:
 

 -W <class name>
  The full class name of the classifier.
  eg: weka.classifiers.bayes.NaiveBayes

 -C <index>
  The index of the class for which IR statistics
  are to be output. (default 1)

 -I <index>
  The index of an attribute to output in the
  results. This attribute should identify an
  instance in order to know which instances are
  in the test set of a cross validation. if 0
  no output (default 0).

 -P
  Add target and prediction columns to the result
  for each fold.

 
 Options specific to classifier weka.classifiers.rules.ZeroR:
 

 -D
  If set, classifier is run in debug mode and
  may output additional info to the console

All options after -- will be passed to the split evaluator.

Specified by:

setOptions in interface OptionHandler

Parameters:

options - the list of options as an array of strings

Throws:

Exception - if an option is not supported

getOptions

public String[] getOptions()

Gets the current settings of the result producer.

Specified by:

getOptions in interface OptionHandler

Returns:

an array of strings suitable for passing to setOptions

toString

public String toString()

Gets a text descrption of the result producer.

Overrides:

toString in class Object

Returns:

a text description of the result producer.

getRevision

public String getRevision()

Returns the revision string.

Specified by:

getRevision in interface RevisionHandler

Returns:

the revision

main

public static void main(String[] args)

Quick test of timestamp

Parameters:

args - the commandline options