All Packages Class Hierarchy This Package Previous Next Index WEKA's home
java.lang.Object | +----weka.attributeSelection.ASSearch | +----weka.attributeSelection.RaceSearch
For more information see:
Moore, A. W. and Lee, M. S. (1994). Efficient algorithms for minimising
cross validation error. Proceedings of the Eleventh International
Conference on Machine Learning. pp 190--198.
Valid options are:
-R
-L
-T
-F
-A
-Q
-N
-J
-Z
-R
-L
-T
-F
-A
-Q
-N
-J
-Z
0 = forward, 1 = backward, 2 = schemata, 3 = rank.
significance level to use for t-tests.
threshold for considering mean errors of two subsets the same
0 = 10 fold, 1 = leave-one-out (selected automatically for schemata race
the attribute evaluator to use when doing a rank search
produce a ranked list of attributes. Selecting this option forces
the race type to be forward. Racing continues until *all* attributes
have been selected, thus producing a ranked list of attributes.
Specify the number of attributes to retain. Overides any threshold.
Use in conjunction with -Q.
Specify a threshold by which the AttributeSelection module can discard
attributes. Use in conjunction with -Q.
Turn on verbose output for monitoring the search
attributeEvaluatorTipText()
debugTipText()
foldsTipText()
generateRankingTipText()
getAttributeEvaluator()
getCalculatedNumToSelect()
getDebug()
getFoldsType()
getGenerateRanking()
getNumToSelect()
getOptions()
getRaceType()
getSelectionThreshold()
getSignificanceLevel()
getThreshold()
globalInfo()
listOptions()
numToSelectTipText()
raceTypeTipText()
rankedAttributes()
search(ASEvaluation, Instances)
selectionThresholdTipText()
setAttributeEvaluator(ASEvaluation)
setDebug(boolean)
setFoldsType(SelectedTag)
setGenerateRanking(boolean)
setNumToSelect(int)
setOptions(String[])
setRaceType(SelectedTag)
setSelectionThreshold(double)
setSignificanceLevel(double)
setThreshold(double)
significanceLevelTipText()
thresholdTipText()
toString()
TAGS_SELECTION
public static final Tag TAGS_SELECTION[]
XVALTAGS_SELECTION
public static final Tag XVALTAGS_SELECTION[]
RaceSearch
public RaceSearch()
globalInfo
public String globalInfo()
raceTypeTipText
public String raceTypeTipText()
setRaceType
public void setRaceType(SelectedTag d)
getRaceType
public SelectedTag getRaceType()
significanceLevelTipText
public String significanceLevelTipText()
setSignificanceLevel
public void setSignificanceLevel(double sig)
getSignificanceLevel
public double getSignificanceLevel()
thresholdTipText
public String thresholdTipText()
setThreshold
public void setThreshold(double t)
getThreshold
public double getThreshold()
foldsTipText
public String foldsTipText()
setFoldsType
public void setFoldsType(SelectedTag d)
getFoldsType
public SelectedTag getFoldsType()
debugTipText
public String debugTipText()
setDebug
public void setDebug(boolean d)
getDebug
public boolean getDebug()
attributeEvaluatorTipText
public String attributeEvaluatorTipText()
setAttributeEvaluator
public void setAttributeEvaluator(ASEvaluation newEvaluator)
getAttributeEvaluator
public ASEvaluation getAttributeEvaluator()
generateRankingTipText
public String generateRankingTipText()
setGenerateRanking
public void setGenerateRanking(boolean doRank)
getGenerateRanking
public boolean getGenerateRanking()
numToSelectTipText
public String numToSelectTipText()
setNumToSelect
public void setNumToSelect(int n)
getNumToSelect
public int getNumToSelect()
getCalculatedNumToSelect
public int getCalculatedNumToSelect()
selectionThresholdTipText
public String selectionThresholdTipText()
setSelectionThreshold
public void setSelectionThreshold(double threshold)
getSelectionThreshold
public double getSelectionThreshold()
listOptions
public Enumeration listOptions()
setOptions
public void setOptions(String options[]) throws Exception
0 = forward, 1 = backward, 2 = schemata, 3 = rank.
significance level to use for t-tests.
threshold for considering mean errors of two subsets the same
0 = 10 fold, 1 = leave-one-out (selected automatically for schemata race
the attribute evaluator to use when doing a rank search
produce a ranked list of attributes. Selecting this option forces
the race type to be forward. Racing continues until *all* attributes
have been selected, thus producing a ranked list of attributes.
Specify the number of attributes to retain. Overides any threshold.
Use in conjunction with -Q.
Specify a threshold by which the AttributeSelection module can discard
attributes. Use in conjunction with -Q.
Turn on verbose output for monitoring the search
getOptions
public String[] getOptions()
search
public int[] search(ASEvaluation ASEval,
Instances data) throws Exception
rankedAttributes
public double[][] rankedAttributes() throws Exception
toString
public String toString()
All Packages Class Hierarchy This Package Previous Next Index WEKA's home