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
|
#ifndef _TARGET_PHRASE_EXTRACTOR_H_
#define _TARGET_PHRASE_EXTRACTOR_H_
#include <memory>
#include <unordered_map>
#include <vector>
using namespace std;
namespace extractor {
typedef vector<pair<int, int>> PhraseAlignment;
class Alignment;
class DataArray;
class Phrase;
class PhraseBuilder;
class RuleExtractorHelper;
class Vocabulary;
class TargetPhraseExtractor {
public:
TargetPhraseExtractor(shared_ptr<DataArray> target_data_array,
shared_ptr<Alignment> alignment,
shared_ptr<PhraseBuilder> phrase_builder,
shared_ptr<RuleExtractorHelper> helper,
shared_ptr<Vocabulary> vocabulary,
int max_rule_span,
bool require_tight_phrases);
virtual ~TargetPhraseExtractor();
// Finds all the target phrases that can extracted from a span in the
// target sentence (matching the given set of target phrase gaps).
virtual vector<pair<Phrase, PhraseAlignment>> ExtractPhrases(
const vector<pair<int, int>>& target_gaps, const vector<int>& target_low,
int target_phrase_low, int target_phrase_high,
const unordered_map<int, int>& source_indexes, int sentence_id) const;
protected:
TargetPhraseExtractor();
private:
// Computes the cartesian product over the sets of possible target phrase
// chunks.
void GeneratePhrases(
vector<pair<Phrase, PhraseAlignment>>& target_phrases,
const vector<pair<int, int>>& ranges, int index,
vector<int>& subpatterns, const vector<int>& target_gap_order,
int target_phrase_low, int target_phrase_high,
const unordered_map<int, int>& source_indexes, int sentence_id) const;
shared_ptr<DataArray> target_data_array;
shared_ptr<Alignment> alignment;
shared_ptr<PhraseBuilder> phrase_builder;
shared_ptr<RuleExtractorHelper> helper;
shared_ptr<Vocabulary> vocabulary;
int max_rule_span;
bool require_tight_phrases;
};
} // namespace extractor
#endif
|
