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
|
/*
* Data.h
* met - Minimum Error Training
*
* Created by Nicola Bertoldi on 13/05/08.
*
*/
#ifndef DATA_H
#define DATA_H
using namespace std;
#include <limits>
#include <vector>
#include <iostream>
#include "Util.h"
#include "FeatureData.h"
#include "ScoreData.h"
class Scorer;
class Data
{
private:
Scorer* theScorer;
std::string score_type;
size_t number_of_scores;
SparseVector sparse_weights;
protected:
// TODO: Use smart pointers for exceptional-safety.
ScoreData* scoredata;
FeatureData* featdata;
public:
explicit Data(Scorer& sc, const std::string& sparseweightsfile="");
~Data();
inline void clear() {
scoredata->clear();
featdata->clear();
}
ScoreData* getScoreData() {
return scoredata;
}
FeatureData* getFeatureData() {
return featdata;
}
Scorer* getScorer() {
return theScorer;
}
inline size_t NumberOfFeatures() const {
return featdata->NumberOfFeatures();
}
inline void NumberOfFeatures(size_t v) {
featdata->NumberOfFeatures(v);
}
inline std::string Features() const {
return featdata->Features();
}
inline void Features(const std::string &f) {
featdata->Features(f);
}
void loadnbest(const std::string &file);
void load(const std::string &featfile,const std::string &scorefile) {
featdata->load(featfile, sparse_weights);
scoredata->load(scorefile);
}
void save(const std::string &featfile,const std::string &scorefile, bool bin=false) {
// if (bin) cerr << "Binary write mode is selected" << endl;
// else cerr << "Binary write mode is NOT selected" << endl;
featdata->save(featfile, bin);
scoredata->save(scorefile, bin);
}
inline bool existsFeatureNames() const {
return featdata->existsFeatureNames();
}
inline std::string getFeatureName(size_t idx) const {
return featdata->getFeatureName(idx);
}
inline size_t getFeatureIndex(const std::string& name) const {
return featdata->getFeatureIndex(name);
}
/**
* Create shard_count shards. If shard_size == 0, then the shards are non-overlapping
* and exhaust the data. If 0 < shard_size <= 1, then shards are chosen by sampling
* the data (with replacement) and shard_size is interpreted as the proportion
* of the total size.
*/
void createShards(size_t shard_count, float shard_size, const std::string& scorerconfig,
std::vector<Data>& shards);
};
#endif // DATA_H
|
