magpie.data.utilities.filters

Class BaseDatasetFilter

java.lang.Object

magpie.data.utilities.filters.BaseDatasetFilter

All Implemented Interfaces:

java.io.Serializable, Options

Direct Known Subclasses:

AllMetalsFilter, BestInAlloySystemFilter, ClassProbabilityFilter, ClassRangeFilter, ComparisonOperatorFilter, CompositionDistanceFilter, CompositionSetDistanceFilter, ContainsElementFilter, EntryRankerFilter, IQROutlierFilter, MultiFilter, PredictionOutlierFilter, PropertyRangeFilter, RegressionErrorFilter, StabilityFilter, ZScoreOutlierFilter

public abstract class BaseDatasetFilter
extends java.lang.Object
implements java.io.Serializable, Options

Abstract model for classes that filter undesired entries out of a dataset. User has the ability to determine whether entries that pass a filter should be removed or kept.

Implementations must fulfill the following operations:

• label(magpie.data.Dataset) Generate a array defining if each entry passes the filter
• train(magpie.data.Dataset) If needed, train before filtering
• Options.setOptions(java.util.List<java.lang.Object>) Set options to fit the user's desires

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
BaseDatasetFilter()

Method Summary

Modifier and Type	Method and Description
void	filter(Dataset D) Filters entries out of dataset
abstract boolean[]	label(Dataset D) Given a dataset, determine which entries passes the filter.
protected boolean[]	parallelLabel(Dataset D) Label entries in parallel.
protected int	parallelMinimum() Minimum number of entries to label in parallel.
void	setExclude(boolean Exclude) Define whether entries that pass should be removed, or whether an entry must pass the filter to be kept.
boolean	toExclude()
abstract void	train(Dataset TrainingSet) Train a dataset splitter, if necessary

Methods inherited from class java.lang.Object

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

Methods inherited from interface magpie.utility.interfaces.Options

printUsage, setOptions

Constructor Detail

BaseDatasetFilter

public BaseDatasetFilter()

Method Detail

setExclude

public void setExclude(boolean Exclude)

Define whether entries that pass should be removed, or whether an entry must pass the filter to be kept.

Parameters:

Exclude - Whether entries that pass a filter should be removed

toExclude

public boolean toExclude()

Returns:

Whether this filter is set to exclude or include entries that pass

parallelMinimum

protected int parallelMinimum()

Minimum number of entries to label in parallel.

Returns:

Implementer-specified minimum threshold

label

public abstract boolean[] label(Dataset D)

Given a dataset, determine which entries passes the filter. Please ensure that D remains unaltered (clone if you must)

Parameters:

D - Dataset to be labeled

Returns:

A list of booleans marking whether an entry passes

parallelLabel

protected boolean[] parallelLabel(Dataset D)

Label entries in parallel.

Parameters:

D - Dataset to be labeled

Returns:

A list of entries to be labeled

filter

public void filter(Dataset D)

Filters entries out of dataset

Parameters:

D - Dataset to filter

train

public abstract void train(Dataset TrainingSet)

Train a dataset splitter, if necessary

Parameters:

TrainingSet - Dataset to use for training