FST::Result_Tracker_Feature_Stats< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET > Class Template Reference

Collects evaluated subsets to eventually provide Dependency-Aware Feature Ranking coefficients. More...

#include <result_tracker_feature_stats.hpp>

Inheritance diagram for FST::Result_Tracker_Feature_Stats< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >:

[legend]

Collaboration diagram for FST::Result_Tracker_Feature_Stats< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >:

[legend]

List of all members.

Classes
struct	FeatureStat
	Structure to gather feature occurence statistics over probe subset evaluations. More...
struct	SubSizeStat
	Structure to gather probe subset cardinality statistics. More...
Public Types
typedef Result_Tracker_Dupless < RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >	parent
typedef parent::ResultRec	ResultRec
typedef ResultRec *	PResultRec
typedef boost::shared_ptr< SUBSET >	PSubset
Public Member Functions
	Result_Tracker_Feature_Stats (const IDXTYPE capacity_limit=0)
	Result_Tracker_Feature_Stats (const Result_Tracker_Feature_Stats &rtfs)
bool	compute_stats (std::ostream &os=std::cout)
bool	print_stats (std::ostream &os=std::cout)
bool	getFirstDAF (RETURNTYPE &value, DIMTYPE &feature, const unsigned int DAFidx=0) const
bool	getNextDAF (RETURNTYPE &value, DIMTYPE &feature, const unsigned int DAFidx=0) const
Result_Tracker_Feature_Stats *	clone () const
Result_Tracker_Feature_Stats *	sharing_clone () const
Result_Tracker_Feature_Stats *	stateless_clone () const
virtual std::ostream &	print (std::ostream &os) const
Protected Types
typedef vector< DIMTYPE >	ORDERTYPE
Protected Attributes
DIMTYPE	_n
vector< FeatureStat >	_stats
vector< SubSizeStat >	_dstat
ORDERTYPE	_order [3]
DIMTYPE	_itersize [3]

---

Detailed Description

template<class RETURNTYPE, typename IDXTYPE, typename DIMTYPE, class SUBSET>
class FST::Result_Tracker_Feature_Stats< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >

Collects evaluated subsets to eventually provide Dependency-Aware Feature Ranking coefficients.

Dependency-Aware Feature ranking (DAF) is a new type of ranking method especially suitable for very-high-dimensional feature selection. Unlike standard individual feature ranking, the DAF ranking reflects "average feature quality in context" and as such is capable of providing significantly better results than BIF (provided the data actually do contain mutually dependent features). The method has been described in combination with Monte Carlo based feature selection permitting Wrapper Criteria even in very-high-dimensional setting. For usage see example35 and example36. For more detailed information see UTIA Technical Report No. 2295.

Examples:

demo34.cpp, and demo35t.cpp.

---

Member Function Documentation

template<class RETURNTYPE, typename IDXTYPE, typename DIMTYPE, class SUBSET>

Result_Tracker_Feature_Stats* FST::Result_Tracker_Feature_Stats< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >::clone

(

)

const [inline, virtual]

create 1:1 independent clone of the current object

Reimplemented from FST::Result_Tracker_Dupless< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >.

Referenced by FST::Result_Tracker_Feature_Stats< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >::stateless_clone().

template<class RETURNTYPE, typename IDXTYPE, typename DIMTYPE, class SUBSET>

Result_Tracker_Feature_Stats* FST::Result_Tracker_Feature_Stats< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >::sharing_clone

(

)

const [inline, virtual]

create equivalent clone of the current object, parmitting read-only access to structures in the source object (allows referencing instead of copying of large memory structures). may be faster and save space but requires more caution with respect to concurrency Use example: Data_Accessor memory data representation cloning

Reimplemented from FST::Result_Tracker_Dupless< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >.

template<class RETURNTYPE , typename IDXTYPE , typename DIMTYPE , class SUBSET >

Result_Tracker_Feature_Stats< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET > * FST::Result_Tracker_Feature_Stats< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >::stateless_clone

(

)

const [inline, virtual]

create clone of the current object, ignoring internal temporary structures to save speed. Does not replicate exact object state. The clone must be used carefully in a way that ensures internal structure re-initialization Use example: Data_Splitter cloning or Classifier_SVM cloning due to inability to clone external structures defined in LibSVM

Reimplemented from FST::Result_Tracker_Dupless< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >.

References FST::Result_Tracker_Feature_Stats< RETURNTYPE, IDXTYPE, DIMTYPE, SUBSET >::clone().

---

The documentation for this class was generated from the following file:

• result_tracker_feature_stats.hpp