freeling::fex Class Reference

Class fex implements a feature extractor. More...

#include <fex.h>

Collaboration diagram for freeling::fex:

List of all members.

Public Member Functions
	fex (const std::wstring &, const std::wstring &, const std::map< std::wstring, const feature_function * > &)
	constructor, given rule file, lexicon file (may be empty), and custom functions
	~fex ()
	destructor
void	encode_name (sentence &, std::vector< std::set< std::wstring > > &) const
	encode given sentence in features as feature names.
void	encode_int (sentence &, std::vector< std::set< int > > &) const
	encode given sentence in features as integer feature codes
void	encode_all (sentence &, std::vector< std::set< std::wstring > > &, std::vector< std::set< int > > &) const
	encode given sentence in features as integer feature codes and as features names
std::vector< std::set < std::wstring > >	encode_name (sentence &) const
	encode given sentence in features as feature names. Return result suitable for Java/perl APIs
std::vector< std::set< int > >	encode_int (sentence &) const
	encode given sentence in features as integer feature codes. Return result apt for Java/perl APIs
void	clear_lexicon ()
	clear lexicon
void	encode_to_lexicon (sentence &)
	encode sentence and add features to current lexicon
void	save_lexicon (const std::wstring &, double) const
	save lexicon to a file, filtering features with low occurrence rate
Private Member Functions
void	get_features (sentence &, std::vector< std::set< std::wstring > > &, std::vector< std::set< int > > &, int) const
	extract features from a sentence
void	read_condition (std::wistringstream &, const std::wstring &, const std::wstring &, std::list< fex_condition > &, int &)
	read rule conditions
void	precompute_once (const fex_rulepack &, sentence &) const
	Apply all rules in the pack once to each word in the sentence.
void	precompute_range (const fex_rulepack &, sentence &) const
	Apply all rules in the pack only to proper range around words matching pack condition.
Private Attributes
const tagset *	Tags
	tagset desription used to compute short versions of PoS tags
std::list< fex_rulepack >	packs
	rules used by the feature extractor
fex_lexicon	lex
	lexicon used to filter features.
std::map< std::wstring, std::set< std::wstring > >	set_files
	loaded set files (loaded once, may be used by several conditions in different rules)
const std::map< std::wstring, const feature_function * > &	feat_functs
	currently loaded specific feature extraction functions

---

Detailed Description

Class fex implements a feature extractor.

---

Constructor & Destructor Documentation

freeling::fex::fex	(	const std::wstring &	,
		const std::wstring &	,
		const std::map< std::wstring, const feature_function * > &
	)

constructor, given rule file, lexicon file (may be empty), and custom functions

freeling::fex::~fex

(

)

destructor

---

Member Function Documentation

void freeling::fex::clear_lexicon

(

)

clear lexicon

void freeling::fex::encode_all	(	sentence &	,
		std::vector< std::set< std::wstring > > &	,
		std::vector< std::set< int > > &
	)		const

encode given sentence in features as integer feature codes and as features names

void freeling::fex::encode_int	(	sentence &	,
		std::vector< std::set< int > > &
	)		const

encode given sentence in features as integer feature codes

Referenced by freeling::bioner::analyze().

std::vector<std::set<int> > freeling::fex::encode_int

(

sentence &

)

const

encode given sentence in features as integer feature codes. Return result apt for Java/perl APIs

void freeling::fex::encode_name	(	sentence &	,
		std::vector< std::set< std::wstring > > &
	)		const

encode given sentence in features as feature names.

std::vector<std::set<std::wstring> > freeling::fex::encode_name

(

sentence &

)

const

encode given sentence in features as feature names. Return result suitable for Java/perl APIs

void freeling::fex::encode_to_lexicon

(

sentence &

)

encode sentence and add features to current lexicon

void freeling::fex::get_features	(	sentence &	,
		std::vector< std::set< std::wstring > > &	,
		std::vector< std::set< int > > &	,
		int
	)		const [private]

extract features from a sentence

void freeling::fex::precompute_once	(	const fex_rulepack &	,
		sentence &
	)		const [private]

Apply all rules in the pack once to each word in the sentence.

void freeling::fex::precompute_range	(	const fex_rulepack &	,
		sentence &
	)		const [private]

Apply all rules in the pack only to proper range around words matching pack condition.

void freeling::fex::read_condition	(	std::wistringstream &	,
		const std::wstring &	,
		const std::wstring &	,
		std::list< fex_condition > &	,
		int &
	)		[private]

read rule conditions

void freeling::fex::save_lexicon	(	const std::wstring &	,
		double
	)		const

save lexicon to a file, filtering features with low occurrence rate

---

Member Data Documentation

const std::map<std::wstring,const feature_function *>& freeling::fex::feat_functs [private]

currently loaded specific feature extraction functions

fex_lexicon freeling::fex::lex [private]

lexicon used to filter features.

std::list<fex_rulepack> freeling::fex::packs [private]

rules used by the feature extractor

std::map<std::wstring, std::set<std::wstring> > freeling::fex::set_files [private]

loaded set files (loaded once, may be used by several conditions in different rules)

const tagset* freeling::fex::Tags [private]

tagset desription used to compute short versions of PoS tags

---

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

• fex.h