Welcome to mirror list, hosted at ThFree Co, Russian Federation.

TypeDef.h « moses - github.com/moses-smt/mosesdecoder.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
blob: a87b38acbf163b55c01add44675ce5df28a57a45 (plain)
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
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
// -*- mode: c++; indent-tabs-mode: nil; tab-width: 2 -*-
/***********************************************************************
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
***********************************************************************/

#pragma once

#include <list>
#include <limits>
#include <vector>
#include <string>
#include <stdint.h>

#include <boost/shared_ptr.hpp>
#include <boost/weak_ptr.hpp>

//! all the typedefs and enums goes here


namespace Moses
{

#define PROJECT_NAME		"moses"

#ifndef BOS_
#define BOS_ "<s>" //Beginning of sentence symbol
#endif
#ifndef EOS_
#define EOS_ "</s>" //End of sentence symbol
#endif

#define UNKNOWN_FACTOR	"UNK"
#define EPSILON         "*EPS*"

#define NOT_FOUND 			std::numeric_limits<size_t>::max()
#define MAX_NGRAM_SIZE  20

const size_t DEFAULT_CUBE_PRUNING_POP_LIMIT = 1000;
const size_t DEFAULT_CUBE_PRUNING_DIVERSITY = 0;
const size_t DEFAULT_MAX_HYPOSTACK_SIZE = 200;
const size_t DEFAULT_MAX_TRANS_OPT_CACHE_SIZE = 10000;
const size_t DEFAULT_MAX_TRANS_OPT_SIZE	= 5000;
const size_t DEFAULT_MAX_PART_TRANS_OPT_SIZE = 10000;
//#ifdef PT_UG
// setting to std::numeric_limits<size_t>::max() makes the regression test for (deprecated) PhraseDictionaryDynamicSuffixArray fail.
// const size_t DEFAULT_MAX_PHRASE_LENGTH = 100000;
//#else
const size_t DEFAULT_MAX_PHRASE_LENGTH = 20;
//#endif
const size_t DEFAULT_MAX_CHART_SPAN			= 20;
const size_t ARRAY_SIZE_INCR					= 10; //amount by which a phrase gets resized when necessary
const float LOWEST_SCORE							= -100.0f;
const float DEFAULT_BEAM_WIDTH				= 0.00001f;
const float DEFAULT_EARLY_DISCARDING_THRESHOLD		= 0.0f;
const float DEFAULT_TRANSLATION_OPTION_THRESHOLD	= 0.0f;
const size_t DEFAULT_VERBOSE_LEVEL = 1;

// output floats with five significant digits
static const size_t PRECISION = 3;

// tolerance for equality in floating point comparisons
const float FLOAT_EPSILON = 0.0001;

// enums.
// must be 0, 1, 2, ..., unless otherwise stated

// can only be 2 at the moment
const int NUM_LANGUAGES = 2;

// Looking for MAX_NUM_FACTORS?  It's defined by the build system: bjam --max-factors=4

enum FactorDirection {
  Input,			//! Source factors
  Output			//! Target factors
};

enum DecodeType {
  Translate,
  Generate
};

namespace LexReorderType
{
enum LexReorderType { // explain values
  Backward,
  Forward,
  Bidirectional,
  Fe,
  F
};
}

namespace DistortionOrientationType
{
enum DistortionOrientationOptions {
  Monotone, //distinguish only between monotone and non-monotone as possible orientations
  Msd //further separate non-monotone into swapped and discontinuous
};
}

enum InputTypeEnum {
  SentenceInput         = 0,
  ConfusionNetworkInput	= 1,
  WordLatticeInput      = 2,
  TreeInputType         = 3,
  //,WordLatticeInput2 = 4,
  TabbedSentenceInput    = 5,
  ForestInputType        = 6
};

enum XmlInputType {
  XmlPassThrough = 0,
  XmlIgnore      = 1,
  XmlExclusive   = 2,
  XmlInclusive   = 3,
  XmlConstraint	 = 4
};

enum DictionaryFind {
  Best		= 0,
  All		= 1
};

// Note: StaticData uses SearchAlgorithm to determine whether the translation
// model is phrase-based or syntax-based.  If you add a syntax-based search
// algorithm here then you should also update StaticData::IsSyntax().
enum SearchAlgorithm {
  Normal = 0,
  CubePruning	= 1,
  //,CubeGrowing = 2
  CYKPlus = 3,
  //NormalBatch  = 4,
  ChartIncremental = 5,
  SyntaxS2T = 6,
  SyntaxT2S = 7,
  SyntaxT2S_SCFG = 8,
  SyntaxF2S = 9,
  DefaultSearchAlgorithm = 777 // means: use StaticData.m_searchAlgorithm
};

enum SourceLabelOverlap {
  SourceLabelOverlapAdd = 0,
  SourceLabelOverlapReplace = 1,
  SourceLabelOverlapDiscard = 2
};

enum WordAlignmentSort {
  NoSort = 0,
  TargetOrder = 1
};

enum FormatType {
  MosesFormat,
  HieroFormat
};

enum S2TParsingAlgorithm {
  RecursiveCYKPlus,
  Scope3
};

// typedef
typedef size_t FactorType;

typedef std::vector<float> Scores;
typedef std::vector<std::string> WordAlignments;

typedef std::vector<FactorType> FactorList;

typedef std::pair<std::vector<std::string const*>,WordAlignments > StringWordAlignmentCand;

class TranslationTask;
typedef boost::shared_ptr<TranslationTask> ttasksptr;
typedef boost::weak_ptr<TranslationTask> ttaskwptr;
}

#define SPTR boost::shared_ptr
#define WPTR boost::weak_ptr