Class mRMR

java.lang.Object

org.tribuo.classification.fs.mRMR

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.config.Configurable, com.oracle.labs.mlrg.olcut.provenance.Provenancable<FeatureSelectorProvenance>, FeatureSelector<Label>

public final class mRMR
extends Object
implements FeatureSelector<Label>

Selects features according to the Minimum Redundancy Maximum Relevance algorithm.

Uses equal width binning for the feature values.

See:

 Peng H, Long F, Ding C.
 "Feature Selection Based on Mutual Information: Criteria of Max-Dependency, Max-Relevance, and Min-Redundancy"
 IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE PAMI), 2005.
 

Field Summary

Fields inherited from interface org.tribuo.FeatureSelector

SELECT_ALL

Constructor Summary

Constructors

Constructor	Description
mRMR(int k, int numBins, int numThreads)	Constructs a mRMR feature selector that ranks the top k features.

Method Summary

Modifier and Type	Method	Description
FeatureSelectorProvenance	getProvenance()
boolean	isOrdered()	Does this feature selection algorithm return an ordered feature set?
void	postConfig()	Used by the OLCUT configuration system, and should not be called by external code.
SelectedFeatureSet	select(Dataset<Label> dataset)	Selects features according to this selection algorithm from the specified dataset.

Methods inherited from class java.lang.Object

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

Constructor Details

mRMR

public mRMR(int k,
 int numBins,
 int numThreads)

Constructs a mRMR feature selector that ranks the top k features.

Continuous features are binned into numBins equal width bins.

Parameters:

k - The number of features to rank.

numBins - The number of bins, must be greater than 1.

numThreads - The number of computation threads to use.

Method Details

postConfig

public void postConfig()

Used by the OLCUT configuration system, and should not be called by external code.

Specified by:

postConfig in interface com.oracle.labs.mlrg.olcut.config.Configurable

isOrdered

public boolean isOrdered()

Description copied from interface: FeatureSelector

Does this feature selection algorithm return an ordered feature set?

Specified by:

isOrdered in interface FeatureSelector<Label>

Returns:

True if the set is ordered.

select

public SelectedFeatureSet select(Dataset<Label> dataset)

Description copied from interface: FeatureSelector

Selects features according to this selection algorithm from the specified dataset.

Specified by:

select in interface FeatureSelector<Label>

Parameters:

dataset - The dataset to use.

Returns:

A selected feature set.

getProvenance

public FeatureSelectorProvenance getProvenance()

Specified by:

getProvenance in interface com.oracle.labs.mlrg.olcut.provenance.Provenancable<FeatureSelectorProvenance>