de.jstacs.classifiers.assessment
Class RepeatedHoldOutAssessParameterSet

java.lang.Object
  de.jstacs.parameters.ParameterSet
      de.jstacs.classifiers.assessment.ClassifierAssessmentAssessParameterSet
          de.jstacs.classifiers.assessment.RepeatedHoldOutAssessParameterSet

All Implemented Interfaces:

GalaxyConvertible, Storable, Cloneable

public class RepeatedHoldOutAssessParameterSet
extends ClassifierAssessmentAssessParameterSet

This class implements a ClassifierAssessmentAssessParameterSet that must be used to call method assess( ... ) of a RepeatedHoldOutExperiment. It contains user specific parameters necessary for a run of a RepeatedHoldOutExperiment.

Author:

Andre Gohr (bioinf (nospam:.) ag (nospam:@) googlemail (nospam:.) com)

Nested Class Summary

Nested classes/interfaces inherited from class de.jstacs.parameters.ParameterSet

ParameterSet.ParameterList

Field Summary

Fields inherited from class de.jstacs.parameters.ParameterSet

errorMessage, parameters, parent

Constructor Summary

RepeatedHoldOutAssessParameterSet()
Constructs a new RepeatedHoldOutAssessParameterSet with empty parameter values.

RepeatedHoldOutAssessParameterSet(DataSet.PartitionMethod dataSplitMethod, int elementLength, boolean exceptionIfMPNotComputable, int repeats, double[] percents)
Constructs a new RepeatedHoldOutAssessParameterSet with given parameter values.

RepeatedHoldOutAssessParameterSet(StringBuffer representation)
The standard constructor for the interface Storable.

Method Summary

Collection<Result>	getAnnotation() Returns a Collection of parameters containing informations about this ClassifierAssessmentAssessParameterSet.
DataSet.PartitionMethod	getDataSplitMethod() Returns the DataSet.PartitionMethod defining how the mutually exclusive random-splits of user supplied data are generated.
protected SimpleParameterSet	getParameterSetContainingASingleDoubleValue(double percent) Creates a new ParameterSet containing a single double-SimpleParameter.
double[]	getPercents() Returns an array containing for each class the percentage of user supplied data that is used in each iteration as test dataset.
int	getRepeats() Returns the repeats defined by this RepeatedHoldOutAssessParameterSet (repeats define how many iterations (train and test classifiers) of that RepeatedHoldOutExperiment this RepeatedHoldOutAssessParameterSet is used with are performed).

Methods inherited from class de.jstacs.classifiers.assessment.ClassifierAssessmentAssessParameterSet

getElementLength, getExceptionIfMPNotComputable, getStoreAll, setStoreAll

Methods inherited from class de.jstacs.parameters.ParameterSet

clone, fromGalaxy, fromXML, getAllParameterNames, getComment, getComment, getErrorMessage, getIndex, getName, getName, getNumberOfParameters, getParameterAt, getParameterForName, getParent, hasDefaultOrIsSet, initParameterList, initParameterList, isAtomic, isComparable, parametersLoaded, reset, setParent, toGalaxy, toXML

Methods inherited from class java.lang.Object

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

Constructor Detail

RepeatedHoldOutAssessParameterSet

public RepeatedHoldOutAssessParameterSet()
                                  throws ParameterException,
                                         CloneNotSupportedException

Constructs a new RepeatedHoldOutAssessParameterSet with empty parameter values. This constructor should only be used to create "filled" RepeatedHoldOutAssessParameterSets, i.e. to create RepeatedHoldOutAssessParameterSets from a set of values and not to fill it from the platform user interface.

Throws:

ParameterException - if the parameters could not be created

CloneNotSupportedException - if the parameter for the percentages could not be created

See Also:

ClassifierAssessmentAssessParameterSet.ClassifierAssessmentAssessParameterSet()

RepeatedHoldOutAssessParameterSet

public RepeatedHoldOutAssessParameterSet(StringBuffer representation)
                                  throws NonParsableException

The standard constructor for the interface Storable. Constructs a RepeatedHoldOutAssessParameterSet out of its XML representation.

Parameters:

representation - the XML representation as StringBuffer

Throws:

NonParsableException - if the RepeatedHoldOutAssessParameterSet could not be reconstructed out of the XML representation (the StringBuffer representation could not be parsed)

See Also:

ClassifierAssessmentAssessParameterSet.ClassifierAssessmentAssessParameterSet(StringBuffer), Storable

RepeatedHoldOutAssessParameterSet

public RepeatedHoldOutAssessParameterSet(DataSet.PartitionMethod dataSplitMethod,
                                         int elementLength,
                                         boolean exceptionIfMPNotComputable,
                                         int repeats,
                                         double[] percents)
                                  throws ParameterException,
                                         CloneNotSupportedException

Constructs a new RepeatedHoldOutAssessParameterSet with given parameter values.

Parameters:

dataSplitMethod - defines the method used to split user supplied data into k mutually exclusive random-splits (available options are: DataSet.PartitionMethod.PARTITION_BY_NUMBER_OF_ELEMENTS and DataSet.PartitionMethod.PARTITION_BY_NUMBER_OF_SYMBOLS)

elementLength - defines the length of elements (sequences) the classifiers to be assessed are able to classify

exceptionIfMPNotComputable - a RepeatedHoldOutAssessParameterSet is used in combination with an AbstractPerformanceMeasure-object to call assess( ... )-methods of RepeatedHoldOutExperiments. If exceptionIfMPNotComputable==true an exception is thrown in case of a user selected measure parameters that could not be computed.

repeats - the number of repeats of each iteration (mutually exclusive, randomly split data to obtain test and train datasets, train classifiers using train datasets and test them using test datasets) of that RepeatedHoldOutExperiment this RepeatedHoldOutAssessParameterSet is used with

percents - this array contains class-wise the percentage of the user supplied data that should be used as test data in each iteration of that RepeatedHoldOutExperiment this RepeatedHoldOutAssessParameterSet is used with

Throws:

ParameterException - if the parameters could not be created

CloneNotSupportedException - if the parameter for the percentages could not be created

See Also:

ClassifierAssessmentAssessParameterSet.ClassifierAssessmentAssessParameterSet(int, boolean), DataSet.PartitionMethod

Method Detail

getParameterSetContainingASingleDoubleValue

protected SimpleParameterSet getParameterSetContainingASingleDoubleValue(double percent)
                                                                  throws SimpleParameter.IllegalValueException

Creates a new ParameterSet containing a single double-SimpleParameter. This ParameterSet is used as a part of the ExpandableParameterSet that contains the test data percent for a specific class.

Parameters:

percent - the double-value to be contained in the returned ParameterSet. If percent==Double.NaN no values are contained in the returned ParameterSet. (The SimpleParameter contained in the returned ParameterSet contains no value).

Returns:

a SimpleParameter

Throws:

SimpleParameter.IllegalValueException - if something went wrong

getRepeats

public int getRepeats()

Returns the repeats defined by this RepeatedHoldOutAssessParameterSet (repeats define how many iterations (train and test classifiers) of that RepeatedHoldOutExperiment this RepeatedHoldOutAssessParameterSet is used with are performed).

Returns:

the repeats defined by this RepeatedHoldOutAssessParameterSet (repeats define how many iterations (train and test classifiers) of that RepeatedHoldOutExperiment this RepeatedHoldOutAssessParameterSet is used with are performed)

getPercents

public double[] getPercents()

Returns an array containing for each class the percentage of user supplied data that is used in each iteration as test dataset.

Returns:

an array containing for each class the percentage of user supplied data that is used in each iteration as test dataset

getDataSplitMethod

public DataSet.PartitionMethod getDataSplitMethod()

Returns the DataSet.PartitionMethod defining how the mutually exclusive random-splits of user supplied data are generated.

Returns:

the DataSet.PartitionMethod defining how the mutually exclusive random-splits of user supplied data are generated

See Also:

DataSet.PartitionMethod

getAnnotation

public Collection<Result> getAnnotation()

Description copied from class: ClassifierAssessmentAssessParameterSet

Returns a Collection of parameters containing informations about this ClassifierAssessmentAssessParameterSet.

Overrides:

getAnnotation in class ClassifierAssessmentAssessParameterSet

Returns:

a Collection of parameters containing informations about this ClassifierAssessmentAssessParameterSet