libsent/src/ngram/ngram_access.c

単語列・クラス列の N-gram 確率を求める [詳細]

#include <sent/stddefs.h>
#include <sent/ngram2.h>

ソースコードを見る。

関数
static NNID	search_ngram_core (NGRAM_INFO *ndata, int n, NNID nid_prev, WORD_ID wkey)
	Search for n-gram tuple.
NNID	search_ngram (NGRAM_INFO *ndata, int n, WORD_ID *w)
	Search for N-tuples.
LOGPROB	ngram_prob (NGRAM_INFO *ndata, int n, WORD_ID *w)
	Get N-gram probability of the last word w_n, given context w_1^n-1.
LOGPROB	uni_prob (NGRAM_INFO *ndata, WORD_ID w)
	Get 1-gram probability of in log10.
static NNID	search_bigram (NGRAM_INFO *ndata, WORD_ID w_context, WORD_ID w)
	Find bi-gram entry.
static LOGPROB	bi_prob_normal (NGRAM_INFO *ndata, WORD_ID w1, WORD_ID w2)
	Get LR bi-gram prob: for LR N-gram
static LOGPROB	bi_prob_additional_oldbin (NGRAM_INFO *ndata, WORD_ID w1, WORD_ID w2)
	Get LR bi-gram prob: for RL N-gram with additional LR 2-gram, in old bingram format.
static LOGPROB	bi_prob_additional (NGRAM_INFO *ndata, WORD_ID w1, WORD_ID w2)
	Get LR bi-gram prob: for RL N-gram with additional LR 2-gram.
static LOGPROB	bi_prob_compute (NGRAM_INFO *ndata, WORD_ID w1, WORD_ID w2)
	Get LR bi-gram prob: for RL N-gram with no LR 2-gram.
LOGPROB	bi_prob (NGRAM_INFO *ndata, WORD_ID w1, WORD_ID w2)
	Get 2-gram probability This function is not used in Julius, since each function of bi_prob_* will be called directly from the search.
void	bi_prob_func_set (NGRAM_INFO *ndata)
	Determinte which bi-gram computation function to be used according to the N-gram type, and set pointer to the proper function into the N-gram data.

---

説明

単語列・クラス列の N-gram 確率を求める

作者:

Akinobu LEE

日付:

Wed Feb 16 07:46:18 2005

Revision

1.5

ngram_access.c で定義されています。

---

関数

static NNID search_ngram_core	(	NGRAM_INFO *	ndata,
		int	n,
		NNID	nid_prev,
		WORD_ID	wkey
	)			[static]

Search for n-gram tuple.

引数:

	ndata	[in] word/class N-gram
	n	[in] N of N-gram
	nid_prev	[in] context (N-1)-gram tuple ID
	wkey	[in] the target word ID

戻り値:

corresponding index to the 2-gram data part if found, or NNID_INVALID if the tuple does not exist in 2-gram.

ngram_access.c の 43 行で定義されています。

参照元 ngram_prob()・search_ngram().

NNID search_ngram	(	NGRAM_INFO *	ndata,
		int	n,
		WORD_ID *	w
	)

Search for N-tuples.

引数:

	ndata	[in] word/class N-gram
	n	[in] N of N-gram (= number of words in w)
	w	[in] word sequence

戻り値:

ngram_access.c の 103 行で定義されています。

参照元 add_bigram().

LOGPROB ngram_prob	(	NGRAM_INFO *	ndata,
		int	n,
		WORD_ID *	w
	)

Get N-gram probability of the last word w_n, given context w_1^n-1.

引数:

	ndata	[in] word/class N-gram
	n	[in] N of N-gram (= number of words in w)
	w	[in] word sequence

戻り値:

ngram_access.c の 135 行で定義されています。

参照元 ngram_forw2back()・ngram_prob().

LOGPROB uni_prob	(	NGRAM_INFO *	ndata,
		WORD_ID	w
	)

Get 1-gram probability of in log10.

引数:

	ndata	[in] word/class N-gram
	w	[in] word/class ID in N-gram

戻り値:

log10 probability .

ngram_access.c の 229 行で定義されています。

参照元 get_nbest_uniprob()・max_successor_prob().

static NNID search_bigram	(	NGRAM_INFO *	ndata,
		WORD_ID	w_context,
		WORD_ID	w
	)			[static]

Find bi-gram entry.

Assumes ct_compaction and is24bit is FALSE on 2-gram

引数:

	ndata	[in] N-gram data that holds the 2-gram
	w_context	[in] context word ID
	w	[in] target word ID

戻り値:

the ID of N-gram tuple entry ID where the (w_context, w) exists.

ngram_access.c の 249 行で定義されています。

参照元 bi_prob_additional()・bi_prob_additional_oldbin()・bi_prob_compute()・bi_prob_normal().

static LOGPROB bi_prob_normal	(	NGRAM_INFO *	ndata,
		WORD_ID	w1,
		WORD_ID	w2
	)			[static]

Get LR bi-gram prob: for LR N-gram

引数:

	ndata	[in] N-gram data that holds the 2-gram
	w1	[in] left context word
	w2	[in] right target word

戻り値:

the log N-gram probability P(w2|w1)

ngram_access.c の 288 行で定義されています。

参照元 bi_prob()・bi_prob_func_set().

static LOGPROB bi_prob_additional_oldbin	(	NGRAM_INFO *	ndata,
		WORD_ID	w1,
		WORD_ID	w2
	)			[static]

Get LR bi-gram prob: for RL N-gram with additional LR 2-gram, in old bingram format.

The old format has 2-gram index in reversed orfer, so this function is for old bingram formats.

引数:

	ndata	[in] N-gram data that holds the 2-gram
	w1	[in] left context word
	w2	[in] right target word

戻り値:

the log N-gram probability P(w2|w1)

ngram_access.c の 320 行で定義されています。

参照元 bi_prob()・bi_prob_func_set().

static LOGPROB bi_prob_additional	(	NGRAM_INFO *	ndata,
		WORD_ID	w1,
		WORD_ID	w2
	)			[static]

Get LR bi-gram prob: for RL N-gram with additional LR 2-gram.

引数:

	ndata	[in] N-gram data that holds the 2-gram
	w1	[in] left context word
	w2	[in] right target word

戻り値:

the log N-gram probability P(w2|w1)

ngram_access.c の 351 行で定義されています。

参照元 bi_prob()・bi_prob_func_set().

static LOGPROB bi_prob_compute	(	NGRAM_INFO *	ndata,
		WORD_ID	w1,
		WORD_ID	w2
	)			[static]

Get LR bi-gram prob: for RL N-gram with no LR 2-gram.

This function will compute the LR 2-gram from the RL 2-gram.

引数:

	ndata	[in] N-gram data that holds the 2-gram
	w1	[in] left context word
	w2	[in] right target word

戻り値:

the log N-gram probability P(w2|w1)

ngram_access.c の 383 行で定義されています。

参照元 bi_prob()・bi_prob_func_set().

LOGPROB bi_prob	(	NGRAM_INFO *	ndata,
		WORD_ID	w1,
		WORD_ID	w2
	)

Get 2-gram probability This function is not used in Julius, since each function of bi_prob_* will be called directly from the search.

引数:

	ndata	[in] N-gram data that holds the 2-gram
	w1	[in] left context word
	w2	[in] right target word

戻り値:

the log N-gram probability P(w2|w1)

ngram_access.c の 419 行で定義されています。

void bi_prob_func_set

(

NGRAM_INFO *

ndata

)

Determinte which bi-gram computation function to be used according to the N-gram type, and set pointer to the proper function into the N-gram data.

引数:

ndata

[i/o] N-gram information to use

ngram_access.c の 449 行で定義されています。

参照元 ngram_read_bin().

---