Package weka.attributeSelection
Class ASSearch
java.lang.Object
weka.attributeSelection.ASSearch
- All Implemented Interfaces:
Serializable
,RevisionHandler
- Direct Known Subclasses:
BestFirst
,ExhaustiveSearch
,GeneticSearch
,GreedyStepwise
,LinearForwardSelection
,RaceSearch
,RandomSearch
,Ranker
,RankSearch
,ScatterSearchV1
,SubsetSizeForwardSelection
Abstract attribute selection search class.
- Version:
- $Revision: 1.12 $
- Author:
- Mark Hall (mhall@cs.waikato.ac.nz)
- See Also:
-
Constructor Summary
Constructors -
Method Summary
Modifier and TypeMethodDescriptionstatic ASSearch
Creates a new instance of a search class given it's class name and (optional) arguments to pass to it's setOptions method.Returns the revision string.static ASSearch[]
makeCopies
(ASSearch model, int num) Creates copies of the current search scheme.abstract int[]
search
(ASEvaluation ASEvaluator, Instances data) Searches the attribute subset/ranking space.
-
Constructor Details
-
ASSearch
public ASSearch()
-
-
Method Details
-
getRevision
Returns the revision string.- Specified by:
getRevision
in interfaceRevisionHandler
- Returns:
- the revision
-
search
Searches the attribute subset/ranking space.- Parameters:
ASEvaluator
- the attribute evaluator to guide the searchdata
- the training instances.- Returns:
- an array (not necessarily ordered) of selected attribute indexes
- Throws:
Exception
- if the search can't be completed
-
forName
Creates a new instance of a search class given it's class name and (optional) arguments to pass to it's setOptions method. If the search method implements OptionHandler and the options parameter is non-null, the search method will have it's options set.- Parameters:
searchName
- the fully qualified class name of the search classoptions
- an array of options suitable for passing to setOptions. May be null.- Returns:
- the newly created search object, ready for use.
- Throws:
Exception
- if the search class name is invalid, or the options supplied are not acceptable to the search class.
-
makeCopies
Creates copies of the current search scheme. Note that this method now uses Serialization to perform a deep copy, so the search object must be fully Serializable. Any currently built model will now be copied as well.- Parameters:
model
- an example search scheme to copynum
- the number of search scheme copies to create.- Returns:
- an array of search schemes.
- Throws:
Exception
- if an error occurs
-