meka.experiment.evaluationstatistics

Interface IncrementalEvaluationStatisticsHandler

All Superinterfaces:

EvaluationStatisticsHandler, LogSupporter, weka.core.OptionHandler, Serializable

All Known Subinterfaces:

OptionalIncrementalEvaluationStatisticsHandler

All Known Implementing Classes:

KeyValuePairs

public interface IncrementalEvaluationStatisticsHandler
extends EvaluationStatisticsHandler

For handlers that support incremental writes.

Version:

$Revision$

Author:

FracPete (fracpete at waikato dot ac dot nz)

Method Summary

Modifier and Type	Method and Description
String	append(List<EvaluationStatistics> stats) Adds the given statistics.
boolean	requires(MultiLabelClassifier classifier, weka.core.Instances dataset) Checks whether the specified combination of classifier and dataset is required for evaluation or already present from previous evaluation.
List<EvaluationStatistics>	retrieve(MultiLabelClassifier classifier, weka.core.Instances dataset) Retrieves the statis for the specified combination of classifier and dataset.
boolean	supportsIncrementalUpdate() Returns whether the handler supports incremental write.

Methods inherited from interface meka.experiment.evaluationstatistics.EvaluationStatisticsHandler

finish, initialize, isThreadSafe, read, write

Methods inherited from interface weka.core.OptionHandler

getOptions, listOptions, setOptions

Methods inherited from interface meka.events.LogSupporter

addLogListener, log, removeLogListener

Method Detail

supportsIncrementalUpdate

boolean supportsIncrementalUpdate()

Returns whether the handler supports incremental write.

Returns:

true if supported

requires

boolean requires(MultiLabelClassifier classifier,
                 weka.core.Instances dataset)

Checks whether the specified combination of classifier and dataset is required for evaluation or already present from previous evaluation.

Parameters:

classifier - the classifier to check

dataset - the dataset to check

Returns:

true if it needs evaluating

retrieve

List<EvaluationStatistics> retrieve(MultiLabelClassifier classifier,
                                    weka.core.Instances dataset)

Retrieves the statis for the specified combination of classifier and dataset.

Parameters:

classifier - the classifier to check

dataset - the dataset to check

Returns:

the stats, null if not available

append

String append(List<EvaluationStatistics> stats)

Adds the given statistics.

Parameters:

stats - the statistics to store

Returns:

null if successfully stored, otherwise error message