mulan.evaluation
Class Evaluator

java.lang.Object
  mulan.evaluation.Evaluator

public class Evaluator
extends Object

Evaluator - responsible for generating evaluation data

Version:

2011.09.06

Author:

rofr, Grigorios Tsoumakas

Constructor Summary

Evaluator()

Method Summary

MultipleEvaluation	crossValidate(MultiLabelLearner learner, MultiLabelInstances data, int someFolds) Evaluates a MultiLabelLearner via cross-validation on given data set with defined number of folds and seed.
MultipleEvaluation	crossValidate(MultiLabelLearner learner, MultiLabelInstances data, List<Measure> measures, int someFolds) Evaluates a MultiLabelLearner via cross-validation on given data set using given evaluation measures with defined number of folds and seed.
Evaluation	evaluate(MultiLabelLearner learner, MultiLabelInstances data) Evaluates a MultiLabelLearner on given test data set.
Evaluation	evaluate(MultiLabelLearner learner, MultiLabelInstances data, List<Measure> measures) Evaluates a MultiLabelLearner on given test data set using specified evaluation measures
void	setSeed(int aSeed) Sets the seed for reproduction of cross-validation results

Methods inherited from class java.lang.Object

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

Constructor Detail

Evaluator

public Evaluator()

Method Detail

setSeed

public void setSeed(int aSeed)

Sets the seed for reproduction of cross-validation results

Parameters:

aSeed - seed for reproduction of cross-validation results

evaluate

public Evaluation evaluate(MultiLabelLearner learner,
                           MultiLabelInstances data,
                           List<Measure> measures)
                    throws IllegalArgumentException,
                           Exception

Evaluates a MultiLabelLearner on given test data set using specified evaluation measures

Parameters:

learner - the learner to be evaluated via cross-validation

data - the data set for cross-validation

measures - the evaluation measures to compute

Returns:

an Evaluation object

Throws:

IllegalArgumentException - if an input parameter is null

Exception

evaluate

public Evaluation evaluate(MultiLabelLearner learner,
                           MultiLabelInstances data)
                    throws IllegalArgumentException,
                           Exception

Evaluates a MultiLabelLearner on given test data set.

Parameters:

learner - the learner to be evaluated

data - the data set for evaluation

Returns:

the evaluation result

Throws:

IllegalArgumentException - if either of input parameters is null.

Exception

crossValidate

public MultipleEvaluation crossValidate(MultiLabelLearner learner,
                                        MultiLabelInstances data,
                                        int someFolds)

Evaluates a MultiLabelLearner via cross-validation on given data set with defined number of folds and seed.

Parameters:

learner - the learner to be evaluated via cross-validation

data - the multi-label data set for cross-validation

someFolds -

Returns:

a MultipleEvaluation object holding the results

crossValidate

public MultipleEvaluation crossValidate(MultiLabelLearner learner,
                                        MultiLabelInstances data,
                                        List<Measure> measures,
                                        int someFolds)

Evaluates a MultiLabelLearner via cross-validation on given data set using given evaluation measures with defined number of folds and seed.

Parameters:

learner - the learner to be evaluated via cross-validation

data - the multi-label data set for cross-validation

measures - the evaluation measures to compute

someFolds -

Returns:

a MultipleEvaluation object holding the results