Feature Selection Toolbox FST4
UTIA · Czech Academy of Sciences
Loading...
Searching...
No Matches
FST::Sequential_Step_Ensemble< RETURNTYPE, DIMTYPE, SUBSET, CRITERION >::SubsetCandidate Class Reference

Nested class to hold feature/subset candidate info in the course of the ensemble voting process. More...

#include <seq_step_ensemble.hpp>

Public Member Functions

 SubsetCandidate (const RETURNTYPE value, const PSubset sub, const DIMTYPE tfeature)
 Constructs a candidate entry, taking a deep (owned) copy of sub – the caller's subset object is mutated/reused across candidates during enumeration, so it cannot be aliased here.
 SubsetCandidate (const SubsetCandidate &fc)
 Copy constructor – unlike the constructor above this SHARES (does not deep-copy) the source's _sub, since by this point the source subset is a stable, privately-owned snapshot rather than a reused enumeration object.

Public Attributes

RETURNTYPE _value
 Dual-purpose: in a member's per-candidate list (_lists) this is the raw value that member criterion returned for _sub; in a _final entry it is instead the accumulated vote (summed dense rank across all members, reset to 0.0 at the start of evaluate_candidates() and filled by its "collect ordering votes" loop) – lower means more preferred.
const PSubset _sub
 owned snapshot of the candidate subset (see the two constructors above)
const DIMTYPE _tempfeature
 copy of sub->getFirstTemporaryFeature() at the time this candidate was constructed – the feature this candidate would add/remove

Detailed Description

template<class RETURNTYPE, typename DIMTYPE, class SUBSET, class CRITERION>
class FST::Sequential_Step_Ensemble< RETURNTYPE, DIMTYPE, SUBSET, CRITERION >::SubsetCandidate

Nested class to hold feature/subset candidate info in the course of the ensemble voting process.


The documentation for this class was generated from the following file:
Copyright © 2000–present ÚTIA, Institute of Information Theory and Automation of the Czech Academy of Sciences · Contacts · This site sets no cookies and uses no analytics.