weka.attributeSelection

Class LFSMethods

java.lang.Object

weka.attributeSelection.LFSMethods

All Implemented Interfaces:

RevisionHandler

public class LFSMethods
extends Object
implements RevisionHandler

Version:

$Revision: 1.3 $

Author:

Martin Guetlein (martin.guetlein@gmail.com)

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
class	LFSMethods.Link2 Class for a node in a linked list.
class	LFSMethods.LinkedList2 Class for handling a linked list.

Constructor Summary

Constructors

Constructor and Description
LFSMethods() empty constructor methods are not static because of access to inner class Link2 and LinkedList2

Method Summary

Modifier and Type	Method and Description
BitSet	floatingForwardSearch(int cacheSize, BitSet startGroup, int[] ranking, int k, boolean incrementK, int maxStale, Instances data, SubsetEvaluator evaluator, boolean verbose) Performs linear floating forward selection ( the stopping criteria cannot be changed to a specific size value )
BitSet	forwardSearch(int cacheSize, BitSet startGroup, int[] ranking, int k, boolean incrementK, int maxStale, int forceResultSize, Instances data, SubsetEvaluator evaluator, boolean verbose) Performs linear forward selection
BitSet	getBestGroup()
BitSet	getBestGroupOfSize(int size)
double	getBestMerit()
int	getNumEvalsCached()
int	getNumEvalsTotal()
String	getRevision() Returns the revision string.
int[]	rankAttributes(Instances data, SubsetEvaluator evaluator, boolean verbose)

Methods inherited from class java.lang.Object

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

Constructor Detail

LFSMethods

public LFSMethods()

empty constructor methods are not static because of access to inner class Link2 and LinkedList2

Method Detail

getBestGroup

public BitSet getBestGroup()

Returns:

best group found by forwardSearch/floatingForwardSearch

getBestMerit

public double getBestMerit()

Returns:

merit of best group found by forwardSearch/floatingForwardSearch

getBestGroupOfSize

public BitSet getBestGroupOfSize(int size)

Returns:

best group of size found by forwardSearch

getNumEvalsCached

public int getNumEvalsCached()

Returns:

number of cached / not performed evaluations

getNumEvalsTotal

public int getNumEvalsTotal()

Returns:

number totally performed evaluations

rankAttributes

public int[] rankAttributes(Instances data,
                            SubsetEvaluator evaluator,
                            boolean verbose)
                     throws Exception

Returns:

ranking (integer array) of attributes in data with evaluator (sorting is NOT stable!)

Throws:

Exception

forwardSearch

public BitSet forwardSearch(int cacheSize,
                            BitSet startGroup,
                            int[] ranking,
                            int k,
                            boolean incrementK,
                            int maxStale,
                            int forceResultSize,
                            Instances data,
                            SubsetEvaluator evaluator,
                            boolean verbose)
                     throws Exception

Performs linear forward selection

Parameters:

cacheSize - chacheSize (times number of instances) to store already evaluated sets

startGroup - start group for search (can be null)

ranking - ranking of attributes (as produced by rankAttributes), no ranking would be [0,1,2,3,4..]

k - number of top k attributes that are taken into account

incrementK - true -> fixed-set, false -> fixed-width

maxStale - number of times the search proceeds even though no improvement was found (1 = hill-climbing)

forceResultSize - stopping criteria changed from no-improvement (forceResultSize=-1) to subset-size

data -

evaluator -

verbose -

Returns:

BitSet, that cotains the best-group found

Throws:

Exception

floatingForwardSearch

public BitSet floatingForwardSearch(int cacheSize,
                                    BitSet startGroup,
                                    int[] ranking,
                                    int k,
                                    boolean incrementK,
                                    int maxStale,
                                    Instances data,
                                    SubsetEvaluator evaluator,
                                    boolean verbose)
                             throws Exception

Performs linear floating forward selection ( the stopping criteria cannot be changed to a specific size value )

Parameters:

cacheSize - chacheSize (times number of instances) to store already evaluated sets

startGroup - start group for search (can be null)

ranking - ranking of attributes (as produced by rankAttributes), no ranking would be [0,1,2,3,4..]

k - number of top k attributes that are taken into account

incrementK - true -> fixed-set, false -> fixed-width

maxStale - number of times the search proceeds even though no improvement was found (1 = hill-climbing)

data -

evaluator -

verbose -

Returns:

BitSet, that cotains the best-group found

Throws:

Exception

getRevision

public String getRevision()

Returns the revision string.

Specified by:

getRevision in interface RevisionHandler

Returns:

the revision