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
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
|
// $Id$
/***********************************************************************
Moses - factored phrase-based language decoder
Copyright (C) 2006 University of Edinburgh
This library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
This library 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
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with this library; if not, write to the Free Software
Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
***********************************************************************/
#include "util/check.hh"
#include <limits>
#include <iostream>
#include <fstream>
#include "LM/SRI.h"
#include "TypeDef.h"
#include "Util.h"
#include "FactorCollection.h"
#include "Phrase.h"
#include "StaticData.h"
#include "Vocab.h"
#include "Ngram.h"
using namespace std;
namespace Moses
{
LanguageModelSRI::LanguageModelSRI()
: m_srilmVocab(0)
, m_srilmModel(0)
{
}
LanguageModelSRI::~LanguageModelSRI()
{
delete m_srilmModel;
delete m_srilmVocab;
}
bool LanguageModelSRI::Load(const std::string &filePath
, FactorType factorType
, size_t nGramOrder)
{
m_srilmVocab = new ::Vocab();
m_srilmModel = new Ngram(*m_srilmVocab, nGramOrder);
m_factorType = factorType;
m_nGramOrder = nGramOrder;
m_filePath = filePath;
m_srilmModel->skipOOVs() = false;
File file( filePath.c_str(), "r" );
m_srilmModel->read(file);
// LM can be ok, just outputs warnings
CreateFactors();
m_unknownId = m_srilmVocab->unkIndex();
return true;
}
void LanguageModelSRI::CreateFactors()
{
// add factors which have srilm id
FactorCollection &factorCollection = FactorCollection::Instance();
std::map<size_t, VocabIndex> lmIdMap;
size_t maxFactorId = 0; // to create lookup vector later on
VocabString str;
VocabIter iter(*m_srilmVocab);
while ( (str = iter.next()) != NULL) {
VocabIndex lmId = GetLmID(str);
size_t factorId = factorCollection.AddFactor(Output, m_factorType, str)->GetId();
lmIdMap[factorId] = lmId;
maxFactorId = (factorId > maxFactorId) ? factorId : maxFactorId;
}
size_t factorId;
m_sentenceStart = factorCollection.AddFactor(Output, m_factorType, BOS_);
factorId = m_sentenceStart->GetId();
lmIdMap[factorId] = GetLmID(BOS_);
maxFactorId = (factorId > maxFactorId) ? factorId : maxFactorId;
m_sentenceStartArray[m_factorType] = m_sentenceStart;
m_sentenceEnd = factorCollection.AddFactor(Output, m_factorType, EOS_);
factorId = m_sentenceEnd->GetId();
lmIdMap[factorId] = GetLmID(EOS_);
maxFactorId = (factorId > maxFactorId) ? factorId : maxFactorId;
m_sentenceEndArray[m_factorType] = m_sentenceEnd;
// add to lookup vector in object
m_lmIdLookup.resize(maxFactorId+1);
fill(m_lmIdLookup.begin(), m_lmIdLookup.end(), m_unknownId);
map<size_t, VocabIndex>::iterator iterMap;
for (iterMap = lmIdMap.begin() ; iterMap != lmIdMap.end() ; ++iterMap) {
m_lmIdLookup[iterMap->first] = iterMap->second;
}
}
VocabIndex LanguageModelSRI::GetLmID( const std::string &str ) const
{
return m_srilmVocab->getIndex( str.c_str(), m_unknownId );
}
VocabIndex LanguageModelSRI::GetLmID( const Factor *factor ) const
{
size_t factorId = factor->GetId();
return ( factorId >= m_lmIdLookup.size()) ? m_unknownId : m_lmIdLookup[factorId];
}
LMResult LanguageModelSRI::GetValue(VocabIndex wordId, VocabIndex *context) const
{
LMResult ret;
ret.score = FloorScore(TransformLMScore(m_srilmModel->wordProb( wordId, context)));
ret.unknown = (wordId == m_unknownId);
return ret;
}
LMResult LanguageModelSRI::GetValue(const vector<const Word*> &contextFactor, State* finalState) const
{
LMResult ret;
FactorType factorType = GetFactorType();
size_t count = contextFactor.size();
if (count <= 0) {
if(finalState)
*finalState = NULL;
ret.score = 0.0;
ret.unknown = false;
return ret;
}
// set up context
VocabIndex ngram[count + 1];
for (size_t i = 0 ; i < count - 1 ; i++) {
ngram[i+1] = GetLmID((*contextFactor[count-2-i])[factorType]);
}
ngram[count] = Vocab_None;
CHECK((*contextFactor[count-1])[factorType] != NULL);
// call sri lm fn
VocabIndex lmId = GetLmID((*contextFactor[count-1])[factorType]);
ret = GetValue(lmId, ngram+1);
if (finalState) {
ngram[0] = lmId;
unsigned int dummy;
*finalState = m_srilmModel->contextID(ngram, dummy);
}
return ret;
}
}
|
