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
|
// $Id$
/***********************************************************************
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 <cassert>
#include <limits>
#include <iostream>
#include <sstream>
#include "LanguageModel.h"
#include "TypeDef.h"
#include "Util.h"
#include "FactorCollection.h"
#include "Phrase.h"
#include "StaticData.h"
using namespace std;
LanguageModel::LanguageModel(bool registerScore, ScoreIndexManager &scoreIndexManager)
{
if (registerScore)
scoreIndexManager.AddScoreProducer(this);
}
LanguageModel::~LanguageModel() {}
// don't inline virtual funcs...
size_t LanguageModel::GetNumScoreComponents() const
{
return 1;
}
void LanguageModel::CalcScore(const Phrase &phrase
, float &fullScore
, float &ngramScore) const
{
fullScore = 0;
ngramScore = 0;
size_t phraseSize = phrase.GetSize();
vector<const Word*> contextFactor;
contextFactor.reserve(m_nGramOrder);
// start of sentence
for (size_t currPos = 0 ; currPos < m_nGramOrder - 1 && currPos < phraseSize ; currPos++)
{
contextFactor.push_back(&phrase.GetWord(currPos));
fullScore += GetValue(contextFactor);
}
if (phraseSize >= m_nGramOrder)
{
contextFactor.push_back(&phrase.GetWord(m_nGramOrder - 1));
ngramScore = GetValue(contextFactor);
}
// main loop
for (size_t currPos = m_nGramOrder; currPos < phraseSize ; currPos++)
{ // used by hypo to speed up lm score calc
for (size_t currNGramOrder = 0 ; currNGramOrder < m_nGramOrder - 1 ; currNGramOrder++)
{
contextFactor[currNGramOrder] = contextFactor[currNGramOrder + 1];
}
contextFactor[m_nGramOrder - 1] = &phrase.GetWord(currPos);
float partScore = GetValue(contextFactor);
ngramScore += partScore;
}
fullScore += ngramScore;
}
LanguageModel::State LanguageModel::GetState(const std::vector<const Word*> &contextFactor, unsigned int* len) const
{
State state;
unsigned int dummy;
if (!len) len = &dummy;
GetValue(contextFactor,&state,len);
return state;
}
|