1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
|
/*
* libpinyin
* Library to deal with pinyin.
*
* Copyright (C) 2011 Peng Wu <alexepico@gmail.com>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
#include <errno.h>
#include <locale.h>
#include "pinyin.h"
#include "k_mixture_model.h"
static guint32 g_prune_k = 3;
static parameter_t g_prune_poss = 0.99;
void print_help(){
printf("Usage: prune_k_mixture_model -k <INT> --CDF <FLOAT> <FILENAME>\n");
}
bool prune_k_mixture_model(KMixtureModelMagicHeader * magic_header,
KMixtureModelSingleGram * & bigram){
bool success;
FlexibleBigramPhraseArray array = g_array_new(FALSE, FALSE, sizeof(KMixtureModelArrayItemWithToken));
bigram->retrieve_all(array);
for ( size_t i = 0; i < array->len; ++i) {
KMixtureModelArrayItemWithToken * item = &g_array_index(array, KMixtureModelArrayItemWithToken, i);
phrase_token_t token = item->m_token;
parameter_t remained_poss = 1;
for ( size_t k = 0; k < g_prune_k; ++k){
remained_poss -= compute_Pr_G_3_with_count
(k, magic_header->m_N, item->m_item.m_WC,
magic_header->m_N - item->m_item.m_N_n_0,
item->m_item.m_n_1);
}
assert(remained_poss >= 0);
if ( remained_poss < g_prune_poss ) {
/* prune this word or phrase. */
KMixtureModelArrayItem removed_item;
bigram->remove_array_item(token, removed_item);
assert( memcmp(&removed_item, &(item->m_item),
sizeof(KMixtureModelArrayItem)) == 0 );
KMixtureModelArrayHeader header;
bigram->get_array_header(header);
guint32 removed_count = removed_item.m_WC;
header.m_WC -= removed_count;
bigram->set_array_header(header);
magic_header->m_WC -= removed_count;
}
}
KMixtureModelArrayHeader header;
bigram->get_array_header(header);
if ( 0 == header.m_WC ){
delete bigram;
bigram = NULL;
}
return true;
}
int main(int argc, char * argv[]){
int i = 1;
const char * bigram_filename = NULL;
setlocale(LC_ALL, "");
while ( i < argc ){
if ( strcmp("--help", argv[i]) == 0 ){
print_help();
exit(0);
} else if ( strcmp("-k", argv[i]) == 0 ){
if ( ++i >= argc ){
print_help();
exit(EINVAL);
}
g_prune_k = atoi(argv[i]);
} else if ( strcmp("--CDF", argv[i]) == 0 ){
if ( ++i >= argc ){
print_help();
exit(EINVAL);
}
g_prune_poss = atof(argv[i]);
} else {
bigram_filename = argv[i];
}
++i;
}
/* TODO: magic header signature check here. */
KMixtureModelBigram bigram(K_MIXTURE_MODEL_MAGIC_NUMBER);
bigram.attach(bigram_filename, ATTACH_READWRITE);
KMixtureModelMagicHeader magic_header;
bigram.get_magic_header(magic_header);
GArray * items = g_array_new(FALSE, FALSE, sizeof(phrase_token_t));
bigram.get_all_items(items);
for ( size_t i = 0; i < items->len; ++i ){
phrase_token_t * token = &g_array_index(items, phrase_token_t, i);
KMixtureModelSingleGram * single_gram = NULL;
bigram.load(*token, single_gram);
prune_k_mixture_model(&magic_header, single_gram);
if ( NULL == single_gram )
bigram.remove(*token);
else bigram.store(*token, single_gram);
}
bigram.set_magic_header(magic_header);
return 0;
}
|
