Collaboration diagram for EST_Ngrammar:

[legend]

Public Types
enum	representation_t { sparse , dense , backoff }

enum	entry_t { frequencies , log_frequencies }

Public Member Functions
	EST_Ngrammar (int o, representation_t r, const EST_StrList &wordlist)

	EST_Ngrammar (int o, representation_t r, const EST_StrList &wordlist, const EST_StrList &predlist)

	EST_Ngrammar (int o, representation_t r, EST_Discrete &v)

void	default_values ()

void	clear ()

bool	init (int o, representation_t r, const EST_StrList &wordlist)

bool	init (int o, representation_t r, const EST_StrList &wordlist, const EST_StrList &predlist)

bool	init (int o, representation_t r, EST_Discrete &v)

bool	init (int o, representation_t r, EST_Discrete &v, EST_Discrete &pv)

int	num_states (void) const

double	samples (void) const

int	order () const

int	get_vocab_length () const

EST_String	get_vocab_word (int i) const

int	get_vocab_word (const EST_String &s) const

int	get_pred_vocab_length () const

EST_String	get_pred_vocab_word (int i) const

int	get_pred_vocab_word (const EST_String &s) const

int	closed_vocab () const

entry_t	entry_type () const

representation_t	representation () const

bool	build (const EST_StrList &filenames, const EST_String &prev=SENTENCE_START_MARKER, const EST_String &prev_prev=SENTENCE_END_MARKER, const EST_String &last=SENTENCE_END_MARKER, const EST_String &input_format="", const EST_String &oov_mode="", const int mincount=1, const int maxcount=10)

void	accumulate (const EST_StrVector &words, const double count=1)

void	accumulate (const EST_IVector &words, const double count=1)

void	make_htk_compatible ()

EST_read_status	load (const EST_String &filename)

EST_read_status	load (const EST_String &filename, const EST_StrList &wordlist)

EST_write_status	save (const EST_String &filename, const EST_String type="cstr_ascii", const bool trace=false, double floor=0.0)

int	wordlist_index (const EST_String &word, const bool report=true) const

const EST_String &	wordlist_index (int i) const

int	predlist_index (const EST_String &word) const

const EST_String &	predlist_index (int i) const

bool	set_entry_type (entry_t new_type)

bool	set_representation (representation_t new_representation)

double	probability (const EST_StrVector &words, bool force=false, const bool trace=false) const

double	frequency (const EST_StrVector &words, bool force=false, const bool trace=false) const

const EST_String &	predict (const EST_StrVector &words, double prob, int state) const

const EST_String &	predict (const EST_StrVector &words) const

const EST_String &	predict (const EST_StrVector &words, double *prob) const

const EST_String &	predict (const EST_IVector &words, double prob, int state) const

const EST_String &	predict (const EST_IVector &words) const

const EST_String &	predict (const EST_IVector &words, double *prob) const

int	find_state_id (const EST_StrVector &words) const

int	find_state_id (const EST_IVector &words) const

int	find_next_state_id (int state, int word) const

double	reverse_probability (const EST_StrVector &words, bool force=false) const

double	reverse_probability (const EST_IVector &words, bool force=false) const

const EST_DiscreteProbDistribution &	prob_dist (const EST_StrVector &words) const

const EST_DiscreteProbDistribution &	prob_dist (const EST_IVector &words) const

const EST_DiscreteProbDistribution &	prob_dist (int state) const

void	fill_window_start (EST_IVector &window, const EST_String &prev, const EST_String &prev_prev) const

void	fill_window_start (EST_StrVector &window, const EST_String &prev, const EST_String &prev_prev) const

bool	ngram_exists (const EST_StrVector &words) const

bool	ngram_exists (const EST_StrVector &words, const double threshold) const

const double	get_backoff_weight (const EST_StrVector &words) const

bool	set_backoff_weight (const EST_StrVector &words, const double w)

void	print_freqs (ostream &os, double floor=0.0)

bool	compute_backoff_weights (const int mincount=1, const int maxcount=10)

bool	merge (EST_Ngrammar &n, float weight)

Protected Member Functions
bool	init_sparse_representation ()

bool	init_dense_representation ()

const double	get_backoff_discount (const int order, const double freq) const

bool	init_backoff_representation ()

void	prune_backoff_representation (EST_BackoffNgrammarState *start_state=NULL)

void	backoff_restore_unigram_states ()

int	find_dense_state_index (const EST_IVector &words, int index=0) const

const EST_StrVector &	make_ngram_from_index (const int i) const

bool	init_vocab (const EST_StrList &wordlist)

bool	init_vocab (const EST_StrList &word_list, const EST_StrList &pred_list)

bool	check_vocab (const EST_StrList &wordlist)

const EST_String &	lastword (const EST_StrVector &words) const

const int	lastword (const EST_IVector &words) const

bool	sparse_to_dense ()

bool	dense_to_sparse ()

void	take_logs ()

void	take_exps ()

void	freqs_to_probs ()

bool	build_sparse (const EST_String &filename, const EST_String &prev, const EST_String &prev_prev, const EST_String &last)

bool	build_ngram (const EST_String &filename, const EST_String &prev, const EST_String &prev_prev, const EST_String &last, const EST_String &input_format)

void	iterate (EST_StrVector &words, void(function)(EST_Ngrammar n, EST_StrVector &words, void params), void params)

void	const_iterate (EST_StrVector &words, void(function)(const EST_Ngrammar const n, EST_StrVector &words, void params), void params) const

bool	p_init (int o, representation_t r)

bool	oov_preprocess (const EST_String &filename, EST_String &new_filename, const EST_String &what)

const EST_NgrammarState &	find_state_const (const EST_StrVector &words) const

EST_NgrammarState &	find_state (const EST_StrVector &words)

const EST_NgrammarState &	find_state_const (const EST_IVector &words) const

EST_NgrammarState &	find_state (const EST_IVector &words)

const EST_DiscreteProbDistribution &	backoff_prob_dist (const EST_StrVector &words) const

const double	backoff_reverse_probability_sub (const EST_StrVector &words, const EST_BackoffNgrammarState *root) const

const double	backoff_probability (const EST_StrVector &words, const bool trace=false) const

const double	backoff_reverse_probability (const EST_StrVector &words) const

const EST_String &	backoff_most_probable (const EST_StrVector &words, double *prob=NULL) const

void	backoff_traverse (EST_BackoffNgrammarState start_state, void(function)(EST_BackoffNgrammarState s, void params), void *params)

void	backoff_traverse (EST_BackoffNgrammarState start_state, void(function)(EST_BackoffNgrammarState s, void params), void *params, const int level)

Protected Attributes
int	p_order

int	p_num_samples

double	p_number_of_sentences

EST_String	p_sentence_start_marker

EST_String	p_sentence_end_marker

representation_t	p_representation

entry_t	p_entry_type

EST_PredictionSuffixTree	sparse_representation

EST_BackoffNgrammarState *	backoff_representation

double	backoff_threshold

double	backoff_unigram_floor_freq

EST_DVector *	backoff_discount

int	p_num_states

EST_NgrammarState *	p_states

EST_Discrete *	vocab

EST_Discrete *	pred_vocab

EST_DiscreteProbDistribution	vocab_pdf

bool	allow_oov

Friends
class	EST_BackoffNgrammar

ostream &	operator<< (ostream &s, EST_Ngrammar &n)

EST_read_status	load_ngram_htk_ascii (const EST_String filename, EST_Ngrammar &n)

EST_read_status	load_ngram_htk_binary (const EST_String filename, EST_Ngrammar &n)

EST_read_status	load_ngram_arpa (const EST_String filename, EST_Ngrammar &n, const EST_StrList &vocab)

EST_read_status	load_ngram_cstr_ascii (const EST_String filename, EST_Ngrammar &n)

EST_read_status	load_ngram_cstr_bin (const EST_String filename, EST_Ngrammar &n)

EST_write_status	save_ngram_htk_ascii_sub (const EST_String &word, ostream *ost, EST_Ngrammar &n, double floor)

EST_write_status	save_ngram_htk_ascii (const EST_String filename, EST_Ngrammar &n, double floor)

EST_write_status	save_ngram_cstr_ascii (const EST_String filename, EST_Ngrammar &n, const bool trace, double floor)

EST_write_status	save_ngram_cstr_bin (const EST_String filename, EST_Ngrammar &n, const bool trace, double floor)

EST_write_status	save_ngram_arpa (const EST_String filename, EST_Ngrammar &n)

EST_write_status	save_ngram_arpa_sub (ostream *ost, EST_Ngrammar &n, const EST_StrVector &words)

EST_write_status	save_ngram_wfst (const EST_String filename, EST_Ngrammar &n)

void	frequency_of_frequencies (EST_DVector &ff, EST_Ngrammar &n, int this_order)

void	map_frequencies (EST_Ngrammar &n, const EST_DVector &map, const int this_order)

bool	Good_Turing_smooth (EST_Ngrammar &n, int maxcount, int mincount)

void	Good_Turing_discount (EST_Ngrammar &ngrammar, const int maxcount, const double default_discount)

void	fs_build_backoff_ngrams (EST_Ngrammar *backoff_ngrams, EST_Ngrammar &ngram)

int	fs_backoff_smooth (EST_Ngrammar *backoff_ngrams, EST_Ngrammar &ngram, int smooth_thresh)

Detailed Description

Definition at line 209 of file EST_Ngrammar.h.

Member Enumeration Documentation

◆ representation_t

enum EST_Ngrammar::representation_t

Definition at line 214 of file EST_Ngrammar.h.

◆ entry_t

enum EST_Ngrammar::entry_t

Definition at line 219 of file EST_Ngrammar.h.

Constructor & Destructor Documentation

◆ EST_Ngrammar() [1/4]

EST_Ngrammar::EST_Ngrammar ( )

inline

Definition at line 366 of file EST_Ngrammar.h.

◆ EST_Ngrammar() [2/4]

EST_Ngrammar::EST_Ngrammar	(	int	o,
		representation_t	r,
		const EST_StrList &	wordlist
	)

inline

Definition at line 368 of file EST_Ngrammar.h.

◆ EST_Ngrammar() [3/4]

EST_Ngrammar::EST_Ngrammar	(	int	o,
		representation_t	r,
		const EST_StrList &	wordlist,
		const EST_StrList &	predlist
	)

inline

Definition at line 375 of file EST_Ngrammar.h.

◆ EST_Ngrammar() [4/4]

EST_Ngrammar::EST_Ngrammar	(	int	o,
		representation_t	r,
		EST_Discrete &	v
	)

inline

Definition at line 382 of file EST_Ngrammar.h.

◆ ~EST_Ngrammar()

EST_Ngrammar::~EST_Ngrammar ( )

Definition at line 498 of file EST_Ngrammar.cc.

Member Function Documentation

◆ init_sparse_representation()

bool EST_Ngrammar::init_sparse_representation ( )

protected

Definition at line 595 of file EST_Ngrammar.cc.

◆ init_dense_representation()

bool EST_Ngrammar::init_dense_representation ( )

protected

Definition at line 575 of file EST_Ngrammar.cc.

◆ get_backoff_discount()

const double EST_Ngrammar::get_backoff_discount	(	const int	order,
		const double	freq
	)		const

protected

Definition at line 2394 of file EST_Ngrammar.cc.

◆ init_backoff_representation()

bool EST_Ngrammar::init_backoff_representation ( )

protected

Definition at line 611 of file EST_Ngrammar.cc.

◆ prune_backoff_representation()

void EST_Ngrammar::prune_backoff_representation ( EST_BackoffNgrammarState * start_state = NULL )

protected

Definition at line 1578 of file EST_Ngrammar.cc.

◆ backoff_restore_unigram_states()

void EST_Ngrammar::backoff_restore_unigram_states ( )

protected

Definition at line 1558 of file EST_Ngrammar.cc.

◆ find_dense_state_index()

int EST_Ngrammar::find_dense_state_index	(	const EST_IVector &	words,
		int	index = `0`
	)		const

protected

Definition at line 1682 of file EST_Ngrammar.cc.

◆ make_ngram_from_index()

const EST_StrVector & EST_Ngrammar::make_ngram_from_index ( const int i ) const

protected

Definition at line 621 of file EST_Ngrammar.cc.

◆ init_vocab() [1/2]

bool EST_Ngrammar::init_vocab ( const EST_StrList & wordlist )

protected

Definition at line 651 of file EST_Ngrammar.cc.

◆ init_vocab() [2/2]

bool EST_Ngrammar::init_vocab	(	const EST_StrList &	word_list,
		const EST_StrList &	pred_list
	)

protected

Definition at line 663 of file EST_Ngrammar.cc.

◆ check_vocab()

bool EST_Ngrammar::check_vocab ( const EST_StrList & wordlist )

protected

Definition at line 677 of file EST_Ngrammar.cc.

◆ lastword() [1/2]

const EST_String & EST_Ngrammar::lastword ( const EST_StrVector & words ) const

inlineprotected

Definition at line 287 of file EST_Ngrammar.h.

◆ lastword() [2/2]

const int EST_Ngrammar::lastword ( const EST_IVector & words ) const

inlineprotected

Definition at line 289 of file EST_Ngrammar.h.

◆ sparse_to_dense()

bool EST_Ngrammar::sparse_to_dense ( )

protected

Definition at line 1668 of file EST_Ngrammar.cc.

◆ dense_to_sparse()

bool EST_Ngrammar::dense_to_sparse ( )

protected

Definition at line 1675 of file EST_Ngrammar.cc.

◆ build_sparse()

bool EST_Ngrammar::build_sparse	(	const EST_String &	filename,
		const EST_String &	prev,
		const EST_String &	prev_prev,
		const EST_String &	last
	)

protected

Definition at line 1020 of file EST_Ngrammar.cc.

◆ build_ngram()

bool EST_Ngrammar::build_ngram	(	const EST_String &	filename,
		const EST_String &	prev,
		const EST_String &	prev_prev,
		const EST_String &	last,
		const EST_String &	input_format
	)

protected

Definition at line 1164 of file EST_Ngrammar.cc.

◆ iterate()

void EST_Ngrammar::iterate	(	EST_StrVector &	words,
		void()(EST_Ngrammar n, EST_StrVector &words, void *params)	function,
		void *	params
	)

protected

Definition at line 2236 of file EST_Ngrammar.cc.

◆ const_iterate()

void EST_Ngrammar::const_iterate	(	EST_StrVector &	words,
		void()(const EST_Ngrammar const n, EST_StrVector &words, void *params)	function,
		void *	params
	)		const

protected

Definition at line 2282 of file EST_Ngrammar.cc.

◆ p_init()

bool EST_Ngrammar::p_init	(	int	o,
		EST_Ngrammar::representation_t	r
	)

protected

Definition at line 539 of file EST_Ngrammar.cc.

◆ oov_preprocess()

bool EST_Ngrammar::oov_preprocess	(	const EST_String &	filename,
		EST_String &	new_filename,
		const EST_String &	what
	)

protected

Definition at line 1052 of file EST_Ngrammar.cc.

◆ find_state_const() [1/2]

const EST_NgrammarState & EST_Ngrammar::find_state_const ( const EST_StrVector & words ) const

protected

Definition at line 1749 of file EST_Ngrammar.cc.

◆ find_state() [1/2]

EST_NgrammarState & EST_Ngrammar::find_state ( const EST_StrVector & words )

protected

Definition at line 1711 of file EST_Ngrammar.cc.

◆ find_state_const() [2/2]

const EST_NgrammarState & EST_Ngrammar::find_state_const ( const EST_IVector & words ) const

protected

Definition at line 1811 of file EST_Ngrammar.cc.

◆ find_state() [2/2]

EST_NgrammarState & EST_Ngrammar::find_state ( const EST_IVector & words )

protected

Definition at line 1785 of file EST_Ngrammar.cc.

◆ backoff_prob_dist()

const EST_DiscreteProbDistribution & EST_Ngrammar::backoff_prob_dist ( const EST_StrVector & words ) const

protected

Definition at line 2369 of file EST_Ngrammar.cc.

◆ backoff_reverse_probability_sub()

const double EST_Ngrammar::backoff_reverse_probability_sub	(	const EST_StrVector &	words,
		const EST_BackoffNgrammarState *	root
	)		const

protected

Definition at line 2531 of file EST_Ngrammar.cc.

◆ backoff_probability()

const double EST_Ngrammar::backoff_probability	(	const EST_StrVector &	words,
		const bool	trace = `false`
	)		const

protected

Definition at line 2409 of file EST_Ngrammar.cc.

◆ backoff_reverse_probability()

const double EST_Ngrammar::backoff_reverse_probability ( const EST_StrVector & words ) const

protected

Definition at line 2604 of file EST_Ngrammar.cc.

◆ backoff_most_probable()

const EST_String & EST_Ngrammar::backoff_most_probable	(	const EST_StrVector &	words,
		double *	prob = `NULL`
	)		const

protected

Definition at line 2628 of file EST_Ngrammar.cc.

◆ backoff_traverse() [1/2]

void EST_Ngrammar::backoff_traverse	(	EST_BackoffNgrammarState *	start_state,
		void()(EST_BackoffNgrammarState s, void *params)	function,
		void *	params
	)

protected

Definition at line 2667 of file EST_Ngrammar.cc.

◆ backoff_traverse() [2/2]

void EST_Ngrammar::backoff_traverse	(	EST_BackoffNgrammarState *	start_state,
		void()(EST_BackoffNgrammarState s, void *params)	function,
		void *	params,
		const int	level
	)

protected

Definition at line 2693 of file EST_Ngrammar.cc.

◆ default_values()

void EST_Ngrammar::default_values ( )

Definition at line 483 of file EST_Ngrammar.cc.

◆ clear()

void EST_Ngrammar::clear ( )

Definition at line 503 of file EST_Ngrammar.cc.

◆ init() [1/4]

bool EST_Ngrammar::init	(	int	o,
		EST_Ngrammar::representation_t	r,
		const EST_StrList &	wordlist
	)

Definition at line 508 of file EST_Ngrammar.cc.

◆ init() [2/4]

bool EST_Ngrammar::init	(	int	o,
		EST_Ngrammar::representation_t	r,
		const EST_StrList &	wordlist,
		const EST_StrList &	predlist
	)

Definition at line 514 of file EST_Ngrammar.cc.

◆ init() [3/4]

bool EST_Ngrammar::init	(	int	o,
		EST_Ngrammar::representation_t	r,
		EST_Discrete &	v
	)

Definition at line 521 of file EST_Ngrammar.cc.

◆ init() [4/4]

bool EST_Ngrammar::init	(	int	o,
		EST_Ngrammar::representation_t	r,
		EST_Discrete &	v,
		EST_Discrete &	pv
	)

Definition at line 530 of file EST_Ngrammar.cc.

◆ num_states()

int EST_Ngrammar::num_states ( void ) const

inline

Definition at line 400 of file EST_Ngrammar.h.

◆ samples()

double EST_Ngrammar::samples ( void ) const

inline

Definition at line 401 of file EST_Ngrammar.h.

◆ order()

int EST_Ngrammar::order ( ) const

inline

Definition at line 402 of file EST_Ngrammar.h.

◆ get_vocab_length()

int EST_Ngrammar::get_vocab_length ( ) const

inline

Definition at line 403 of file EST_Ngrammar.h.

◆ get_vocab_word() [1/2]

EST_String EST_Ngrammar::get_vocab_word ( int i ) const

Definition at line 1989 of file EST_Ngrammar.cc.

◆ get_vocab_word() [2/2]

int EST_Ngrammar::get_vocab_word ( const EST_String & s ) const

Definition at line 1997 of file EST_Ngrammar.cc.

◆ get_pred_vocab_length()

int EST_Ngrammar::get_pred_vocab_length ( ) const

inline

Definition at line 406 of file EST_Ngrammar.h.

◆ get_pred_vocab_word() [1/2]

EST_String EST_Ngrammar::get_pred_vocab_word ( int i ) const

inline

Definition at line 407 of file EST_Ngrammar.h.

◆ get_pred_vocab_word() [2/2]

int EST_Ngrammar::get_pred_vocab_word ( const EST_String & s ) const

inline

Definition at line 408 of file EST_Ngrammar.h.

◆ closed_vocab()

int EST_Ngrammar::closed_vocab ( ) const

inline

Definition at line 410 of file EST_Ngrammar.h.

◆ entry_type()

entry_t EST_Ngrammar::entry_type ( ) const

inline

Definition at line 411 of file EST_Ngrammar.h.

◆ representation()

representation_t EST_Ngrammar::representation ( ) const

inline

Definition at line 412 of file EST_Ngrammar.h.

◆ build()

bool EST_Ngrammar::build	(	const EST_StrList &	filenames,
		const EST_String &	prev = `SENTENCE_START_MARKER`,
		const EST_String &	prev_prev = `SENTENCE_END_MARKER`,
		const EST_String &	last = `SENTENCE_END_MARKER`,
		const EST_String &	input_format = `""`,
		const EST_String &	oov_mode = `""`,
		const int	mincount = `1`,
		const int	maxcount = `10`
	)

Definition at line 759 of file EST_Ngrammar.cc.

◆ accumulate() [1/2]

void EST_Ngrammar::accumulate	(	const EST_StrVector &	words,
		const double	count = `1`
	)

Definition at line 884 of file EST_Ngrammar.cc.

◆ accumulate() [2/2]

void EST_Ngrammar::accumulate	(	const EST_IVector &	words,
		const double	count = `1`
	)

Definition at line 914 of file EST_Ngrammar.cc.

◆ make_htk_compatible()

void EST_Ngrammar::make_htk_compatible ( )

Definition at line 2205 of file EST_Ngrammar.cc.

◆ load() [1/2]

EST_read_status EST_Ngrammar::load ( const EST_String & filename )

Definition at line 2121 of file EST_Ngrammar.cc.

◆ load() [2/2]

EST_read_status EST_Ngrammar::load	(	const EST_String &	filename,
		const EST_StrList &	wordlist
	)

Definition at line 2160 of file EST_Ngrammar.cc.

◆ save()

EST_write_status EST_Ngrammar::save	(	const EST_String &	filename,
		const EST_String	type = `"cstr_ascii"`,
		const bool	trace = `false`,
		double	floor = `0.0`
	)

Definition at line 2213 of file EST_Ngrammar.cc.

◆ wordlist_index() [1/2]

int EST_Ngrammar::wordlist_index	(	const EST_String &	word,
		const bool	report = `true`
	)		const

Definition at line 733 of file EST_Ngrammar.cc.

◆ wordlist_index() [2/2]

const EST_String & EST_Ngrammar::wordlist_index ( int i ) const

Definition at line 697 of file EST_Ngrammar.cc.

◆ predlist_index() [1/2]

int EST_Ngrammar::predlist_index ( const EST_String & word ) const

Definition at line 702 of file EST_Ngrammar.cc.

◆ predlist_index() [2/2]

const EST_String & EST_Ngrammar::predlist_index ( int i ) const

Definition at line 728 of file EST_Ngrammar.cc.

◆ set_entry_type()

bool EST_Ngrammar::set_entry_type ( EST_Ngrammar::entry_t new_type )

Definition at line 1657 of file EST_Ngrammar.cc.

◆ set_representation()

bool EST_Ngrammar::set_representation ( EST_Ngrammar::representation_t new_representation )

Definition at line 1836 of file EST_Ngrammar.cc.

◆ probability()

double EST_Ngrammar::probability	(	const EST_StrVector &	words,
		bool	force = `false`,
		const bool	trace = `false`
	)		const

Definition at line 1853 of file EST_Ngrammar.cc.

◆ frequency()

double EST_Ngrammar::frequency	(	const EST_StrVector &	words,
		bool	force = `false`,
		const bool	trace = `false`
	)		const

Definition at line 1876 of file EST_Ngrammar.cc.

◆ predict() [1/6]

const EST_String & EST_Ngrammar::predict	(	const EST_StrVector &	words,
		double *	prob,
		int *	state
	)		const

Definition at line 1899 of file EST_Ngrammar.cc.

◆ predict() [2/6]

const EST_String & EST_Ngrammar::predict ( const EST_StrVector & words ) const

inline

Definition at line 463 of file EST_Ngrammar.h.

◆ predict() [3/6]

const EST_String & EST_Ngrammar::predict	(	const EST_StrVector &	words,
		double *	prob
	)		const

inline

Definition at line 465 of file EST_Ngrammar.h.

◆ predict() [4/6]

const EST_String & EST_Ngrammar::predict	(	const EST_IVector &	words,
		double *	prob,
		int *	state
	)		const

Definition at line 1927 of file EST_Ngrammar.cc.

◆ predict() [5/6]

const EST_String & EST_Ngrammar::predict ( const EST_IVector & words ) const

inline

Definition at line 469 of file EST_Ngrammar.h.

◆ predict() [6/6]

const EST_String & EST_Ngrammar::predict	(	const EST_IVector &	words,
		double *	prob
	)		const

inline

Definition at line 471 of file EST_Ngrammar.h.

◆ find_state_id() [1/2]

int EST_Ngrammar::find_state_id ( const EST_StrVector & words ) const

Definition at line 1955 of file EST_Ngrammar.cc.

◆ find_state_id() [2/2]

int EST_Ngrammar::find_state_id ( const EST_IVector & words ) const

Definition at line 1972 of file EST_Ngrammar.cc.

◆ find_next_state_id()

int EST_Ngrammar::find_next_state_id	(	int	state,
		int	word
	)		const

Definition at line 1699 of file EST_Ngrammar.cc.

◆ reverse_probability() [1/2]

double EST_Ngrammar::reverse_probability	(	const EST_StrVector &	words,
		bool	force = `false`
	)		const

Definition at line 2003 of file EST_Ngrammar.cc.

◆ reverse_probability() [2/2]

double EST_Ngrammar::reverse_probability	(	const EST_IVector &	words,
		bool	force = `false`
	)		const

Definition at line 2032 of file EST_Ngrammar.cc.

◆ prob_dist() [1/3]

const EST_DiscreteProbDistribution & EST_Ngrammar::prob_dist ( const EST_StrVector & words ) const

Definition at line 2070 of file EST_Ngrammar.cc.

◆ prob_dist() [2/3]

const EST_DiscreteProbDistribution & EST_Ngrammar::prob_dist ( const EST_IVector & words ) const

Definition at line 2095 of file EST_Ngrammar.cc.

◆ prob_dist() [3/3]

const EST_DiscreteProbDistribution & EST_Ngrammar::prob_dist ( int state ) const

Definition at line 2064 of file EST_Ngrammar.cc.

◆ fill_window_start() [1/2]

void EST_Ngrammar::fill_window_start	(	EST_IVector &	window,
		const EST_String &	prev,
		const EST_String &	prev_prev
	)		const

Definition at line 1030 of file EST_Ngrammar.cc.

◆ fill_window_start() [2/2]

void EST_Ngrammar::fill_window_start	(	EST_StrVector &	window,
		const EST_String &	prev,
		const EST_String &	prev_prev
	)		const

Definition at line 1041 of file EST_Ngrammar.cc.

◆ ngram_exists() [1/2]

bool EST_Ngrammar::ngram_exists ( const EST_StrVector & words ) const

Definition at line 956 of file EST_Ngrammar.cc.

◆ ngram_exists() [2/2]

bool EST_Ngrammar::ngram_exists	(	const EST_StrVector &	words,
		const double	threshold
	)		const

Definition at line 984 of file EST_Ngrammar.cc.

◆ get_backoff_weight()

const double EST_Ngrammar::get_backoff_weight ( const EST_StrVector & words ) const

Definition at line 997 of file EST_Ngrammar.cc.

◆ set_backoff_weight()

bool EST_Ngrammar::set_backoff_weight	(	const EST_StrVector &	words,
		const double	w
	)

Definition at line 1008 of file EST_Ngrammar.cc.

◆ print_freqs()

void EST_Ngrammar::print_freqs	(	ostream &	os,
		double	floor = `0.0`
	)

Definition at line 2328 of file EST_Ngrammar.cc.

◆ compute_backoff_weights()

bool EST_Ngrammar::compute_backoff_weights	(	const int	mincount = `1`,
		const int	maxcount = `10`
	)

Definition at line 1489 of file EST_Ngrammar.cc.

◆ merge()

bool EST_Ngrammar::merge	(	EST_Ngrammar &	n,
		float	weight
	)

Definition at line 2739 of file EST_Ngrammar.cc.

Friends And Related Function Documentation

◆ EST_BackoffNgrammar

friend class EST_BackoffNgrammar

friend

Definition at line 583 of file EST_Ngrammar.h.

◆ operator<<

ostream & operator<<	(	ostream &	s,
		EST_Ngrammar &	n
	)

friend

Definition at line 1629 of file EST_Ngrammar.cc.

◆ load_ngram_htk_ascii

EST_read_status load_ngram_htk_ascii	(	const EST_String	filename,
		EST_Ngrammar &	n
	)

friend

Definition at line 54 of file ngrammar_io.cc.

◆ load_ngram_htk_binary

EST_read_status load_ngram_htk_binary	(	const EST_String	filename,
		EST_Ngrammar &	n
	)

friend

Definition at line 62 of file ngrammar_io.cc.

◆ load_ngram_arpa

EST_read_status load_ngram_arpa	(	const EST_String	filename,
		EST_Ngrammar &	n,
		const EST_StrList &	vocab
	)

friend

Definition at line 70 of file ngrammar_io.cc.

◆ load_ngram_cstr_ascii

EST_read_status load_ngram_cstr_ascii	(	const EST_String	filename,
		EST_Ngrammar &	n
	)

friend

Definition at line 218 of file ngrammar_io.cc.

◆ load_ngram_cstr_bin

EST_read_status load_ngram_cstr_bin	(	const EST_String	filename,
		EST_Ngrammar &	n
	)

friend

Definition at line 280 of file ngrammar_io.cc.

◆ save_ngram_htk_ascii_sub

EST_write_status save_ngram_htk_ascii_sub	(	const EST_String &	word,
		ostream *	ost,
		EST_Ngrammar &	n,
		double	floor
	)

friend

Definition at line 407 of file ngrammar_io.cc.

◆ save_ngram_htk_ascii

EST_write_status save_ngram_htk_ascii	(	const EST_String	filename,
		EST_Ngrammar &	n,
		double	floor
	)

friend

Definition at line 548 of file ngrammar_io.cc.

◆ save_ngram_cstr_ascii

EST_write_status save_ngram_cstr_ascii	(	const EST_String	filename,
		EST_Ngrammar &	n,
		const bool	trace,
		double	floor
	)

friend

Definition at line 729 of file ngrammar_io.cc.

◆ save_ngram_cstr_bin

EST_write_status save_ngram_cstr_bin	(	const EST_String	filename,
		EST_Ngrammar &	n,
		const bool	trace,
		double	floor
	)

friend

Definition at line 826 of file ngrammar_io.cc.

◆ save_ngram_arpa

EST_write_status save_ngram_arpa	(	const EST_String	filename,
		EST_Ngrammar &	n
	)

friend

Definition at line 653 of file ngrammar_io.cc.

◆ save_ngram_wfst

EST_write_status save_ngram_wfst	(	const EST_String	filename,
		EST_Ngrammar &	n
	)

friend

Definition at line 788 of file ngrammar_io.cc.

◆ frequency_of_frequencies

void frequency_of_frequencies	(	EST_DVector &	ff,
		EST_Ngrammar &	n,
		int	this_order
	)

friend

Definition at line 205 of file ngrammar_aux.cc.

◆ map_frequencies

void map_frequencies	(	EST_Ngrammar &	n,
		const EST_DVector &	map,
		const int	this_order
	)

friend

Definition at line 304 of file ngrammar_aux.cc.

◆ Good_Turing_smooth

bool Good_Turing_smooth	(	EST_Ngrammar &	n,
		int	maxcount,
		int	mincount
	)

friend

Definition at line 408 of file ngrammar_aux.cc.

◆ Good_Turing_discount

void Good_Turing_discount	(	EST_Ngrammar &	ngrammar,
		const int	maxcount,
		const double	default_discount
	)

friend

Definition at line 539 of file ngrammar_aux.cc.

◆ fs_build_backoff_ngrams

void fs_build_backoff_ngrams	(	EST_Ngrammar *	backoff_ngrams,
		EST_Ngrammar &	ngram
	)

friend

Definition at line 73 of file freqsmooth.cc.

◆ fs_backoff_smooth

int fs_backoff_smooth	(	EST_Ngrammar *	backoff_ngrams,
		EST_Ngrammar &	ngram,
		int	smooth_thresh
	)

friend

Definition at line 109 of file freqsmooth.cc.

Member Data Documentation

◆ p_order

int EST_Ngrammar::p_order

protected

Definition at line 225 of file EST_Ngrammar.h.

◆ p_num_samples

int EST_Ngrammar::p_num_samples

protected

Definition at line 226 of file EST_Ngrammar.h.

◆ p_number_of_sentences

double EST_Ngrammar::p_number_of_sentences

protected

Definition at line 228 of file EST_Ngrammar.h.

◆ p_sentence_start_marker

EST_String EST_Ngrammar::p_sentence_start_marker

protected

Definition at line 231 of file EST_Ngrammar.h.

◆ p_sentence_end_marker

EST_String EST_Ngrammar::p_sentence_end_marker

protected

Definition at line 232 of file EST_Ngrammar.h.

◆ p_representation

representation_t EST_Ngrammar::p_representation

protected

Definition at line 235 of file EST_Ngrammar.h.

◆ p_entry_type

entry_t EST_Ngrammar::p_entry_type

protected

Definition at line 236 of file EST_Ngrammar.h.

◆ sparse_representation

EST_PredictionSuffixTree EST_Ngrammar::sparse_representation

protected

Definition at line 240 of file EST_Ngrammar.h.

◆ backoff_representation

EST_BackoffNgrammarState* EST_Ngrammar::backoff_representation

protected

Definition at line 250 of file EST_Ngrammar.h.

◆ backoff_threshold

double EST_Ngrammar::backoff_threshold

protected

Definition at line 252 of file EST_Ngrammar.h.

◆ backoff_unigram_floor_freq

double EST_Ngrammar::backoff_unigram_floor_freq

protected

Definition at line 255 of file EST_Ngrammar.h.

◆ backoff_discount

EST_DVector* EST_Ngrammar::backoff_discount

protected

Definition at line 262 of file EST_Ngrammar.h.

◆ p_num_states

int EST_Ngrammar::p_num_states

protected

Definition at line 268 of file EST_Ngrammar.h.

◆ p_states

EST_NgrammarState* EST_Ngrammar::p_states

protected

Definition at line 269 of file EST_Ngrammar.h.

◆ vocab

EST_Discrete* EST_Ngrammar::vocab

protected

Definition at line 276 of file EST_Ngrammar.h.

◆ pred_vocab

EST_Discrete* EST_Ngrammar::pred_vocab

protected

Definition at line 277 of file EST_Ngrammar.h.

◆ vocab_pdf

EST_DiscreteProbDistribution EST_Ngrammar::vocab_pdf

protected

Definition at line 285 of file EST_Ngrammar.h.

◆ allow_oov

bool EST_Ngrammar::allow_oov

protected

Definition at line 292 of file EST_Ngrammar.h.

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

include/EST_Ngrammar.h
grammar/ngram/EST_Ngrammar.cc

Public Types

Public Member Functions

Protected Member Functions

Protected Attributes

Friends

Detailed Description

Member Enumeration Documentation

◆ representation_t

◆ entry_t

Constructor & Destructor Documentation

◆ EST_Ngrammar() [1/4]

◆ EST_Ngrammar() [2/4]

◆ EST_Ngrammar() [3/4]

◆ EST_Ngrammar() [4/4]

◆ ~EST_Ngrammar()

Member Function Documentation

◆ init_sparse_representation()

◆ init_dense_representation()

◆ get_backoff_discount()

◆ init_backoff_representation()

◆ prune_backoff_representation()

◆ backoff_restore_unigram_states()

◆ find_dense_state_index()

◆ make_ngram_from_index()

◆ init_vocab() [1/2]

◆ init_vocab() [2/2]

◆ check_vocab()

◆ lastword() [1/2]

◆ lastword() [2/2]

◆ sparse_to_dense()

◆ dense_to_sparse()

◆ build_sparse()

◆ build_ngram()

◆ iterate()

◆ const_iterate()

◆ p_init()

◆ oov_preprocess()

◆ find_state_const() [1/2]

◆ find_state() [1/2]

◆ find_state_const() [2/2]

◆ find_state() [2/2]

◆ backoff_prob_dist()

◆ backoff_reverse_probability_sub()

◆ backoff_probability()

◆ backoff_reverse_probability()

◆ backoff_most_probable()

◆ backoff_traverse() [1/2]

◆ backoff_traverse() [2/2]

◆ default_values()

◆ clear()

◆ init() [1/4]

◆ init() [2/4]

◆ init() [3/4]

◆ init() [4/4]

◆ num_states()

◆ samples()

◆ order()

◆ get_vocab_length()

◆ get_vocab_word() [1/2]

◆ get_vocab_word() [2/2]

◆ get_pred_vocab_length()

◆ get_pred_vocab_word() [1/2]

◆ get_pred_vocab_word() [2/2]

◆ closed_vocab()

◆ entry_type()

◆ representation()

◆ build()

◆ accumulate() [1/2]

◆ accumulate() [2/2]

◆ make_htk_compatible()

◆ load() [1/2]

◆ load() [2/2]

◆ save()

◆ wordlist_index() [1/2]

◆ wordlist_index() [2/2]

◆ predlist_index() [1/2]

◆ predlist_index() [2/2]

◆ set_entry_type()

◆ set_representation()

◆ probability()