uk.ac.ed.inf.enzml.mulan.learn
Class MulanCrossExperimenter

java.lang.Object
  uk.ac.ed.inf.utils.Initialised
      uk.ac.ed.inf.enzml.mulan.learn.MulanCrossExperimenter

All Implemented Interfaces:

uk.ac.ed.inf.utils.Initialisable

Direct Known Subclasses:

TrainTestExperimenter

public class MulanCrossExperimenter
extends uk.ac.ed.inf.utils.Initialised

Using Mulan (multi-label machine learning), this class initialises and executes either Train-Test evaluation (if two different arff records are given) (calling MulanTrainTestEvaluator) or cross-evaluations (calling MulanCrossEvaluator) if only one arff record is given. Initialisation includes verifying input (arff and xml files) and checking output locations (for saving results to file and database).

Version:

14 Jun 2010

Author:

Luna De Ferrari luna.deferrari-at-ed.ac.uk

Field Summary

Fields inherited from class uk.ac.ed.inf.utils.Initialised

m_initialisationIsCorrect

Constructor Summary

MulanCrossExperimenter(java.lang.String resultsDbPropsFile, java.lang.String arffRecordDbProps, int trainArffRecordId, MulanLearner learner, java.lang.String resultsFolder)
public MulanCrossExperimenter(String resultsDbPropsFile, String arffRecordDbProps, MulanArffRecord trainArffRecord, MulanLearner learner, String resultsFolder) { m_trainArffRecord = trainArffRecord; this.initialise(resultsDbPropsFile, arffRecordDbProps, learner, resultsFolder); }

Method Summary

java.util.Vector<uk.ac.ed.inf.utils.database.TableRow>	crossEvaluateAndSaveResults(int folds) Cross validates and saves results to database
uk.ac.ed.inf.utils.database.DbManager	getExperimentDbManager() public MulanCrossEvaluator getEvaluator() { return m_evaluator; }
java.lang.String	getExperimentPropsFile()
MulanLearner	getLearner() public String getExperimentPropsPath() { return m_experimentPropsFile; }
ResultsSaver	getResultsSaver()
MulanArffRecord	getTrainArffRecord()
static void	main(java.lang.String[] args)
static void	run(int[] arffFilesIds, java.lang.String resultsDbPropsPath) Run a cross-evaluation of several ARFF files (using one or several algorithms, as needed)
ResultsSaver	setResultsSaver()

Methods inherited from class uk.ac.ed.inf.utils.Initialised

initialisationIsCorrect, setInitialisationIsCorrect

Methods inherited from class java.lang.Object

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

Constructor Detail

MulanCrossExperimenter

public MulanCrossExperimenter(java.lang.String resultsDbPropsFile,
                              java.lang.String arffRecordDbProps,
                              int trainArffRecordId,
                              MulanLearner learner,
                              java.lang.String resultsFolder)

public MulanCrossExperimenter(String resultsDbPropsFile, String arffRecordDbProps, MulanArffRecord trainArffRecord, MulanLearner learner, String resultsFolder) { m_trainArffRecord = trainArffRecord; this.initialise(resultsDbPropsFile, arffRecordDbProps, learner, resultsFolder); }

Method Detail

crossEvaluateAndSaveResults

public java.util.Vector<uk.ac.ed.inf.utils.database.TableRow> crossEvaluateAndSaveResults(int folds)

Cross validates and saves results to database

getExperimentDbManager

public uk.ac.ed.inf.utils.database.DbManager getExperimentDbManager()

public MulanCrossEvaluator getEvaluator() { return m_evaluator; }

getExperimentPropsFile

public java.lang.String getExperimentPropsFile()

getLearner

public MulanLearner getLearner()

public String getExperimentPropsPath() { return m_experimentPropsFile; }

getResultsSaver

public ResultsSaver getResultsSaver()

getTrainArffRecord

public MulanArffRecord getTrainArffRecord()

setResultsSaver

public ResultsSaver setResultsSaver()

main

public static void main(java.lang.String[] args)

run

public static void run(int[] arffFilesIds,
                       java.lang.String resultsDbPropsPath)

Run a cross-evaluation of several ARFF files (using one or several algorithms, as needed)

Parameters:

arffFilesIds - a list of ARFF file identifiers in the "arff" database table