weka.classifiers.meta.multisearch

Class AbstractEvaluationFactory<M extends AbstractEvaluationMetrics,W extends AbstractEvaluationWrapper,T extends AbstractEvaluationTask,E>

java.lang.Object

weka.classifiers.meta.multisearch.AbstractEvaluationFactory<M,W,T,E>

All Implemented Interfaces:

Serializable

Direct Known Subclasses:

DefaultEvaluationFactory

public abstract class AbstractEvaluationFactory<M extends AbstractEvaluationMetrics,W extends AbstractEvaluationWrapper,T extends AbstractEvaluationTask,E>
extends Object
implements Serializable

Factory for creating evaluation-related objects.

Version:

$Revision$

Author:

FracPete (fracpete at waikato dot ac dot nz)

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
AbstractEvaluationFactory()

Method Summary

Modifier and Type	Method and Description
abstract M	newMetrics() Returns a new metrics instance.
abstract T	newTask(MultiSearchCapable owner, weka.core.Instances train, weka.core.Instances test, SetupGenerator generator, Point<Object> values, int folds, int eval, int classLabel) Returns a new task.
abstract W	newWrapper(E eval) Returns a new wrapper.

Methods inherited from class java.lang.Object

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

Constructor Detail

AbstractEvaluationFactory

public AbstractEvaluationFactory()

Method Detail

newMetrics

public abstract M newMetrics()

Returns a new metrics instance.

Returns:

the metrics

newWrapper

public abstract W newWrapper(E eval)

Returns a new wrapper.

Parameters:

eval - the evaluation to wrap

Returns:

the wrapper

newTask

public abstract T newTask(MultiSearchCapable owner,
                          weka.core.Instances train,
                          weka.core.Instances test,
                          SetupGenerator generator,
                          Point<Object> values,
                          int folds,
                          int eval,
                          int classLabel)

Returns a new task.

Parameters:

owner - the owning search

train - the training data

test - the test data

generator - the generator

values - the values

folds - the number of folds

eval - the evaluation

classLabel - the class label index (0-based; if applicable)

Returns:

the task