weka.classifiers.meta.multisearch
Class Performance

java.lang.Object
  weka.classifiers.meta.multisearch.Performance

All Implemented Interfaces:

Serializable

public class Performance
extends Object
implements Serializable

A helper class for storing the performance of values in the parameter space. Can be sorted with the PerformanceComparator class.

Version:

$Revision: 4521 $

Author:

fracpete (fracpete at waikato dot ac dot nz)

See Also:

PerformanceComparator, Serialized Form

Field Summary

static int	EVALUATION_ACC evaluation via: Accuracy.
static int	EVALUATION_CC evaluation via: Correlation coefficient.
static int	EVALUATION_COMBINED evaluation via: Combined = (1-CC) + RRSE + RAE.
static int	EVALUATION_KAPPA evaluation via: Kappa statistic.
static int	EVALUATION_MAE evaluation via: Mean absolute error.
static int	EVALUATION_RAE evaluation via: Relative absolute error.
static int	EVALUATION_RMSE evaluation via: Root mean squared error.
static int	EVALUATION_RRSE evaluation via: Root relative squared error.
protected double	m_ACC the Accuracy.
protected double	m_CC the Correlation coefficient.
protected int	m_Evaluation the evaluation type.
protected double	m_Kappa the Kappa statistic.
protected double	m_MAE the Mean absolute error.
protected double	m_RAE the Relative absolute error.
protected double	m_RMSE the Root mean squared error.
protected double	m_RRSE the Root relative squared error.
protected Point<Object>	m_Values the values the filter/classifier were built with.

Constructor Summary

Performance(Point<Object> values, weka.classifiers.Evaluation evaluation, int evalType)
Initializes the performance container.

Method Summary

int	getEvaluation() Returns the evaluation type.
double	getPerformance() returns the performance measure.
double	getPerformance(int evaluation) returns the performance measure.
Point<Object>	getValues() returns the values for this performance.
String	toString() returns a string representation of this performance object.

Methods inherited from class java.lang.Object

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

Field Detail

EVALUATION_CC

public static final int EVALUATION_CC

evaluation via: Correlation coefficient.

See Also:

Constant Field Values

EVALUATION_RMSE

public static final int EVALUATION_RMSE

evaluation via: Root mean squared error.

See Also:

Constant Field Values

EVALUATION_RRSE

public static final int EVALUATION_RRSE

evaluation via: Root relative squared error.

See Also:

Constant Field Values

EVALUATION_MAE

public static final int EVALUATION_MAE

evaluation via: Mean absolute error.

See Also:

Constant Field Values

EVALUATION_RAE

public static final int EVALUATION_RAE

evaluation via: Relative absolute error.

See Also:

Constant Field Values

EVALUATION_COMBINED

public static final int EVALUATION_COMBINED

evaluation via: Combined = (1-CC) + RRSE + RAE.

See Also:

Constant Field Values

EVALUATION_ACC

public static final int EVALUATION_ACC

evaluation via: Accuracy.

See Also:

Constant Field Values

EVALUATION_KAPPA

public static final int EVALUATION_KAPPA

evaluation via: Kappa statistic.

See Also:

Constant Field Values

m_Values

protected Point<Object> m_Values

the values the filter/classifier were built with.

m_Evaluation

protected int m_Evaluation

the evaluation type.

m_CC

protected double m_CC

the Correlation coefficient.

m_RMSE

protected double m_RMSE

the Root mean squared error.

m_RRSE

protected double m_RRSE

the Root relative squared error.

m_MAE

protected double m_MAE

the Mean absolute error.

m_RAE

protected double m_RAE

the Relative absolute error.

m_ACC

protected double m_ACC

the Accuracy.

m_Kappa

protected double m_Kappa

the Kappa statistic.

Constructor Detail

Performance

public Performance(Point<Object> values,
                   weka.classifiers.Evaluation evaluation,
                   int evalType)
            throws Exception

Initializes the performance container. If the Evaluation object is null, then the worst possible values for the measures are assumed (in order to assure a low ranking).

Parameters:

values - the values

evaluation - the evaluation to extract the performance measures from, can be null

evalType - the type of evaluation

Throws:

Exception - if retrieving of measures fails

Method Detail

getEvaluation

public int getEvaluation()

Returns the evaluation type.

Returns:

the type of evaluation

See Also:

MultiSearch.TAGS_EVALUATION

getPerformance

public double getPerformance()

returns the performance measure.

Returns:

the performance measure

getPerformance

public double getPerformance(int evaluation)

returns the performance measure.

Parameters:

evaluation - the type of evaluation to return

Returns:

the performance measure

getValues

public Point<Object> getValues()

returns the values for this performance.

Returns:

the values

toString

public String toString()

returns a string representation of this performance object.

Overrides:

toString in class Object

Returns:

a string representation