#include <ngram2.h>
__ngram_info__のコラボレーション図
変数 | |
| int | n |
| N-gram order (ex. 3 for 3-gram) | |
| int | dir |
| direction (either DIR_LR or DIR_RL) | |
| boolean | from_bin |
| TRUE if source was bingram, otherwise ARPA | |
| boolean | bigram_index_reversed |
| TRUE if read from old (<=3.5.3) bingram, in which case the 2-gram tuple index is reversed (DIR_LR) against the RL 3-gram. | |
| boolean | bos_eos_swap |
| TRUE if swap BOS and SOS on backward N-gram | |
| WORD_ID | max_word_num |
| N-gram vocabulary size | |
| char ** | wname |
| List of word strings. | |
| PATNODE * | root |
| Root of index tree to search n-gram word ID from its name | |
| WORD_ID | unk_id |
| Word ID of unknown word. | |
| int | unk_num |
| Number of dictionary words that are not in this N-gram vocabulary | |
| LOGPROB | unk_num_log |
| Log10 value of unk_num, used for calculating probability of unknown words | |
| boolean | isopen |
| TRUE if dictionary has unknown words, which does not appear in this N-gram | |
| NGRAM_TUPLE_INFO | d [MAX_N] |
| Main body of N-gram info | |
| LOGPROB * | bo_wt_1 |
| back-off weights for 2-gram on 1st pass | |
| LOGPROB * | p_2 |
| 2-gram prob for the 1st pass | |
| LOGPROB(* | bigram_prob )(struct __ngram_info__ *, WORD_ID, WORD_ID) |
| Pointer of a function to compite bigram probability on the 1st pass. See bi_prob_func_set() for details | |
bigrams and trigrams are stored in the form of sequential lists. They are grouped by the same context, and referred from the context ((N-1)-gram) data by the beginning ID and its number.
1.5.1