From 252fb164c208ec8f3005f8a652eb3b48c0644e3d Mon Sep 17 00:00:00 2001 From: Paul Baltescu Date: Fri, 1 Feb 2013 16:11:10 +0000 Subject: Second working commit. --- extractor/features/max_lex_target_given_source.cc | 30 +++++++++++++++++++++++ 1 file changed, 30 insertions(+) create mode 100644 extractor/features/max_lex_target_given_source.cc (limited to 'extractor/features/max_lex_target_given_source.cc') diff --git a/extractor/features/max_lex_target_given_source.cc b/extractor/features/max_lex_target_given_source.cc new file mode 100644 index 00000000..d82182fe --- /dev/null +++ b/extractor/features/max_lex_target_given_source.cc @@ -0,0 +1,30 @@ +#include "max_lex_target_given_source.h" + +#include + +#include "../translation_table.h" + +MaxLexTargetGivenSource::MaxLexTargetGivenSource( + shared_ptr table) : + table(table) {} + +double MaxLexTargetGivenSource::Score(const FeatureContext& context) const { + // TODO(pauldb): Add NULL to source_words, after fixing translation table. + vector source_words = context.source_phrase.GetWords(); + vector target_words = context.target_phrase.GetWords(); + + double score = 0; + for (string target_word: target_words) { + double max_score = 0; + for (string source_word: source_words) { + max_score = max(max_score, + table->GetTargetGivenSourceScore(source_word, target_word)); + } + score += max_score > 0 ? -log10(max_score) : MAX_SCORE; + } + return score; +} + +string MaxLexTargetGivenSource::GetName() const { + return "MaxLexEGivenF"; +} -- cgit v1.2.3 From 9a026ba2db8fa7723374109e6a4a8dcaff8733cd Mon Sep 17 00:00:00 2001 From: Paul Baltescu Date: Thu, 14 Feb 2013 23:17:15 +0000 Subject: Working version of the grammar extractor. --- extractor/Makefile.am | 90 ++- extractor/alignment.cc | 6 +- extractor/alignment.h | 7 +- extractor/alignment_test.cc | 31 + extractor/binary_search_merger.cc | 6 +- extractor/binary_search_merger.h | 6 +- extractor/data_array.cc | 19 +- extractor/data_array.h | 16 +- extractor/features/count_source_target_test.cc | 32 ++ extractor/features/feature.cc | 2 + extractor/features/feature.h | 10 +- extractor/features/is_source_singleton.cc | 2 +- extractor/features/is_source_singleton_test.cc | 35 ++ extractor/features/is_source_target_singleton.cc | 2 +- .../features/is_source_target_singleton_test.cc | 35 ++ extractor/features/max_lex_source_given_target.cc | 5 +- .../features/max_lex_source_given_target_test.cc | 74 +++ extractor/features/max_lex_target_given_source.cc | 5 +- .../features/max_lex_target_given_source_test.cc | 74 +++ extractor/features/sample_source_count.cc | 2 +- extractor/features/sample_source_count_test.cc | 36 ++ extractor/features/target_given_source_coherent.cc | 4 +- .../features/target_given_source_coherent_test.cc | 35 ++ extractor/grammar.cc | 19 +- extractor/grammar.h | 4 + extractor/grammar_extractor.cc | 45 +- extractor/grammar_extractor.h | 8 +- extractor/grammar_extractor_test.cc | 49 ++ extractor/intersector.cc | 31 +- extractor/intersector.h | 16 +- extractor/intersector_test.cc | 6 +- extractor/linear_merger.cc | 12 +- extractor/linear_merger.h | 6 +- extractor/matchings_finder.cc | 4 + extractor/matchings_finder.h | 8 +- extractor/matchings_trie.cc | 14 +- extractor/matchings_trie.h | 5 +- extractor/mocks/mock_alignment.h | 10 + extractor/mocks/mock_binary_search_merger.h | 4 +- extractor/mocks/mock_data_array.h | 4 + extractor/mocks/mock_feature.h | 9 + extractor/mocks/mock_intersector.h | 11 + extractor/mocks/mock_linear_merger.h | 2 +- extractor/mocks/mock_matchings_finder.h | 9 + extractor/mocks/mock_rule_extractor.h | 12 + extractor/mocks/mock_rule_extractor_helper.h | 78 +++ extractor/mocks/mock_rule_factory.h | 9 + extractor/mocks/mock_sampler.h | 9 + extractor/mocks/mock_scorer.h | 10 + extractor/mocks/mock_target_phrase_extractor.h | 12 + extractor/mocks/mock_translation_table.h | 9 + extractor/phrase_builder.cc | 5 +- extractor/phrase_location.cc | 6 +- extractor/precomputation.cc | 8 +- extractor/precomputation.h | 5 +- extractor/rule_extractor.cc | 618 +++++--------------- extractor/rule_extractor.h | 92 ++- extractor/rule_extractor_helper.cc | 356 ++++++++++++ extractor/rule_extractor_helper.h | 82 +++ extractor/rule_extractor_helper_test.cc | 622 +++++++++++++++++++++ extractor/rule_extractor_test.cc | 166 ++++++ extractor/rule_factory.cc | 84 ++- extractor/rule_factory.h | 22 +- extractor/rule_factory_test.cc | 98 ++++ extractor/run_extractor.cc | 9 +- extractor/sample_alignment.txt | 2 + extractor/sampler.cc | 7 +- extractor/sampler.h | 7 +- extractor/scorer.cc | 4 + extractor/scorer.h | 9 +- extractor/scorer_test.cc | 47 ++ extractor/suffix_array.cc | 9 +- extractor/suffix_array_test.cc | 29 +- extractor/target_phrase_extractor.cc | 144 +++++ extractor/target_phrase_extractor.h | 56 ++ extractor/target_phrase_extractor_test.cc | 116 ++++ extractor/translation_table.cc | 47 +- extractor/translation_table.h | 23 +- extractor/translation_table_test.cc | 82 +++ extractor/vocabulary.h | 3 +- 80 files changed, 3007 insertions(+), 700 deletions(-) create mode 100644 extractor/alignment_test.cc create mode 100644 extractor/features/count_source_target_test.cc create mode 100644 extractor/features/is_source_singleton_test.cc create mode 100644 extractor/features/is_source_target_singleton_test.cc create mode 100644 extractor/features/max_lex_source_given_target_test.cc create mode 100644 extractor/features/max_lex_target_given_source_test.cc create mode 100644 extractor/features/sample_source_count_test.cc create mode 100644 extractor/features/target_given_source_coherent_test.cc create mode 100644 extractor/grammar_extractor_test.cc create mode 100644 extractor/mocks/mock_alignment.h create mode 100644 extractor/mocks/mock_feature.h create mode 100644 extractor/mocks/mock_intersector.h create mode 100644 extractor/mocks/mock_matchings_finder.h create mode 100644 extractor/mocks/mock_rule_extractor.h create mode 100644 extractor/mocks/mock_rule_extractor_helper.h create mode 100644 extractor/mocks/mock_rule_factory.h create mode 100644 extractor/mocks/mock_sampler.h create mode 100644 extractor/mocks/mock_scorer.h create mode 100644 extractor/mocks/mock_target_phrase_extractor.h create mode 100644 extractor/mocks/mock_translation_table.h create mode 100644 extractor/rule_extractor_helper.cc create mode 100644 extractor/rule_extractor_helper.h create mode 100644 extractor/rule_extractor_helper_test.cc create mode 100644 extractor/rule_extractor_test.cc create mode 100644 extractor/rule_factory_test.cc create mode 100644 extractor/sample_alignment.txt create mode 100644 extractor/scorer_test.cc create mode 100644 extractor/target_phrase_extractor.cc create mode 100644 extractor/target_phrase_extractor.h create mode 100644 extractor/target_phrase_extractor_test.cc create mode 100644 extractor/translation_table_test.cc (limited to 'extractor/features/max_lex_target_given_source.cc') diff --git a/extractor/Makefile.am b/extractor/Makefile.am index ded06239..c82fc1ae 100644 --- a/extractor/Makefile.am +++ b/extractor/Makefile.am @@ -1,8 +1,17 @@ bin_PROGRAMS = compile run_extractor noinst_PROGRAMS = \ + alignment_test \ binary_search_merger_test \ data_array_test \ + feature_count_source_target_test \ + feature_is_source_singleton_test \ + feature_is_source_target_singleton_test \ + feature_max_lex_source_given_target_test \ + feature_max_lex_target_given_source_test \ + feature_sample_source_count_test \ + feature_target_given_source_coherent_test \ + grammar_extractor_test \ intersector_test \ linear_merger_test \ matching_comparator_test \ @@ -10,27 +19,66 @@ noinst_PROGRAMS = \ matchings_finder_test \ phrase_test \ precomputation_test \ + rule_extractor_helper_test \ + rule_extractor_test \ + rule_factory_test \ sampler_test \ + scorer_test \ suffix_array_test \ + target_phrase_extractor_test \ + translation_table_test \ veb_test -TESTS = sampler_test -#TESTS = binary_search_merger_test \ -# data_array_test \ -# intersector_test \ -# linear_merger_test \ -# matching_comparator_test \ -# matching_test \ -# matchings_finder_test \ -# phrase_test \ -# precomputation_test \ -# suffix_array_test \ -# veb_test +TESTS = alignment_test \ + binary_search_merger_test \ + data_array_test \ + feature_count_source_target_test \ + feature_is_source_singleton_test \ + feature_is_source_target_singleton_test \ + feature_max_lex_source_given_target_test \ + feature_max_lex_target_given_source_test \ + feature_sample_source_count_test \ + feature_target_given_source_coherent_test \ + grammar_extractor_test \ + intersector_test \ + linear_merger_test \ + matching_comparator_test \ + matching_test \ + matchings_finder_test \ + phrase_test \ + precomputation_test \ + rule_extractor_helper_test \ + rule_extractor_test \ + rule_factory_test \ + sampler_test \ + scorer_test \ + suffix_array_test \ + target_phrase_extractor_test \ + translation_table_test \ + veb_test +alignment_test_SOURCES = alignment_test.cc +alignment_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) libextractor.a binary_search_merger_test_SOURCES = binary_search_merger_test.cc binary_search_merger_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a data_array_test_SOURCES = data_array_test.cc data_array_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) libextractor.a +feature_count_source_target_test_SOURCES = features/count_source_target_test.cc +feature_count_source_target_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) libextractor.a +feature_is_source_singleton_test_SOURCES = features/is_source_singleton_test.cc +feature_is_source_singleton_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) libextractor.a +feature_is_source_target_singleton_test_SOURCES = features/is_source_target_singleton_test.cc +feature_is_source_target_singleton_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) libextractor.a +feature_max_lex_source_given_target_test_SOURCES = features/max_lex_source_given_target_test.cc +feature_max_lex_source_given_target_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a +feature_max_lex_target_given_source_test_SOURCES = features/max_lex_target_given_source_test.cc +feature_max_lex_target_given_source_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a +feature_sample_source_count_test_SOURCES = features/sample_source_count_test.cc +feature_sample_source_count_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) libextractor.a +feature_target_given_source_coherent_test_SOURCES = features/target_given_source_coherent_test.cc +feature_target_given_source_coherent_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) libextractor.a +grammar_extractor_test_SOURCES = grammar_extractor_test.cc +grammar_extractor_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a intersector_test_SOURCES = intersector_test.cc intersector_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a linear_merger_test_SOURCES = linear_merger_test.cc @@ -45,10 +93,22 @@ phrase_test_SOURCES = phrase_test.cc phrase_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a precomputation_test_SOURCES = precomputation_test.cc precomputation_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a -suffix_array_test_SOURCES = suffix_array_test.cc -suffix_array_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a +rule_extractor_helper_test_SOURCES = rule_extractor_helper_test.cc +rule_extractor_helper_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a +rule_extractor_test_SOURCES = rule_extractor_test.cc +rule_extractor_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a +rule_factory_test_SOURCES = rule_factory_test.cc +rule_factory_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a sampler_test_SOURCES = sampler_test.cc sampler_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a +scorer_test_SOURCES = scorer_test.cc +scorer_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a +suffix_array_test_SOURCES = suffix_array_test.cc +suffix_array_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a +target_phrase_extractor_test_SOURCES = target_phrase_extractor_test.cc +target_phrase_extractor_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a +translation_table_test_SOURCES = translation_table_test.cc +translation_table_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) $(GMOCK_LDFLAGS) $(GMOCK_LIBS) libextractor.a veb_test_SOURCES = veb_test.cc veb_test_LDADD = $(GTEST_LDFLAGS) $(GTEST_LIBS) libextractor.a @@ -93,10 +153,12 @@ libextractor_a_SOURCES = \ precomputation.cc \ rule.cc \ rule_extractor.cc \ + rule_extractor_helper.cc \ rule_factory.cc \ sampler.cc \ scorer.cc \ suffix_array.cc \ + target_phrase_extractor.cc \ translation_table.cc \ veb.cc \ veb_bitset.cc \ diff --git a/extractor/alignment.cc b/extractor/alignment.cc index 2fa0abac..ff39d484 100644 --- a/extractor/alignment.cc +++ b/extractor/alignment.cc @@ -31,7 +31,11 @@ Alignment::Alignment(const string& filename) { alignments.shrink_to_fit(); } -const vector >& Alignment::GetLinks(int sentence_index) const { +Alignment::Alignment() {} + +Alignment::~Alignment() {} + +vector > Alignment::GetLinks(int sentence_index) const { return alignments[sentence_index]; } diff --git a/extractor/alignment.h b/extractor/alignment.h index 290d6015..f7e79585 100644 --- a/extractor/alignment.h +++ b/extractor/alignment.h @@ -13,10 +13,15 @@ class Alignment { public: Alignment(const string& filename); - const vector >& GetLinks(int sentence_index) const; + virtual vector > GetLinks(int sentence_index) const; void WriteBinary(const fs::path& filepath); + virtual ~Alignment(); + + protected: + Alignment(); + private: vector > > alignments; }; diff --git a/extractor/alignment_test.cc b/extractor/alignment_test.cc new file mode 100644 index 00000000..1bc51a56 --- /dev/null +++ b/extractor/alignment_test.cc @@ -0,0 +1,31 @@ +#include + +#include +#include + +#include "alignment.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class AlignmentTest : public Test { + protected: + virtual void SetUp() { + alignment = make_shared("sample_alignment.txt"); + } + + shared_ptr alignment; +}; + +TEST_F(AlignmentTest, TestGetLinks) { + vector > expected_links = { + make_pair(0, 0), make_pair(1, 1), make_pair(2, 2) + }; + EXPECT_EQ(expected_links, alignment->GetLinks(0)); + expected_links = {make_pair(1, 0), make_pair(2, 1)}; + EXPECT_EQ(expected_links, alignment->GetLinks(1)); +} + +} // namespace diff --git a/extractor/binary_search_merger.cc b/extractor/binary_search_merger.cc index 43d2f734..c1b86a77 100644 --- a/extractor/binary_search_merger.cc +++ b/extractor/binary_search_merger.cc @@ -25,8 +25,10 @@ BinarySearchMerger::~BinarySearchMerger() {} void BinarySearchMerger::Merge( vector& locations, const Phrase& phrase, const Phrase& suffix, - vector::iterator prefix_start, vector::iterator prefix_end, - vector::iterator suffix_start, vector::iterator suffix_end, + const vector::iterator& prefix_start, + const vector::iterator& prefix_end, + const vector::iterator& suffix_start, + const vector::iterator& suffix_end, int prefix_subpatterns, int suffix_subpatterns) const { if (IsIntersectionVoid(prefix_start, prefix_end, suffix_start, suffix_end, prefix_subpatterns, suffix_subpatterns, suffix)) { diff --git a/extractor/binary_search_merger.h b/extractor/binary_search_merger.h index ffa47c8e..c887e012 100644 --- a/extractor/binary_search_merger.h +++ b/extractor/binary_search_merger.h @@ -24,8 +24,10 @@ class BinarySearchMerger { virtual void Merge( vector& locations, const Phrase& phrase, const Phrase& suffix, - vector::iterator prefix_start, vector::iterator prefix_end, - vector::iterator suffix_start, vector::iterator suffix_end, + const vector::iterator& prefix_start, + const vector::iterator& prefix_end, + const vector::iterator& suffix_start, + const vector::iterator& suffix_end, int prefix_subpatterns, int suffix_subpatterns) const; static double BAEZA_YATES_FACTOR; diff --git a/extractor/data_array.cc b/extractor/data_array.cc index 383b08a7..1097caf3 100644 --- a/extractor/data_array.cc +++ b/extractor/data_array.cc @@ -10,9 +10,9 @@ namespace fs = boost::filesystem; using namespace std; -int DataArray::END_OF_FILE = 0; +int DataArray::NULL_WORD = 0; int DataArray::END_OF_LINE = 1; -string DataArray::END_OF_FILE_STR = "__END_OF_FILE__"; +string DataArray::NULL_WORD_STR = "__NULL__"; string DataArray::END_OF_LINE_STR = "__END_OF_LINE__"; DataArray::DataArray() { @@ -47,9 +47,9 @@ DataArray::DataArray(const string& filename, const Side& side) { } void DataArray::InitializeDataArray() { - word2id[END_OF_FILE_STR] = END_OF_FILE; - id2word.push_back(END_OF_FILE_STR); - word2id[END_OF_LINE_STR] = END_OF_FILE; + word2id[NULL_WORD_STR] = NULL_WORD; + id2word.push_back(NULL_WORD_STR); + word2id[END_OF_LINE_STR] = END_OF_LINE; id2word.push_back(END_OF_LINE_STR); } @@ -87,6 +87,10 @@ int DataArray::AtIndex(int index) const { return data[index]; } +string DataArray::GetWordAtIndex(int index) const { + return id2word[data[index]]; +} + int DataArray::GetSize() const { return data.size(); } @@ -103,6 +107,11 @@ int DataArray::GetSentenceStart(int position) const { return sentence_start[position]; } +int DataArray::GetSentenceLength(int sentence_id) const { + // Ignore end of line markers. + return sentence_start[sentence_id + 1] - sentence_start[sentence_id] - 1; +} + int DataArray::GetSentenceId(int position) const { return sentence_id[position]; } diff --git a/extractor/data_array.h b/extractor/data_array.h index 19fbff88..7c120b3c 100644 --- a/extractor/data_array.h +++ b/extractor/data_array.h @@ -2,14 +2,13 @@ #define _DATA_ARRAY_H_ #include -#include +#include #include #include namespace fs = boost::filesystem; using namespace std; -using namespace tr1; enum Side { SOURCE, @@ -18,9 +17,9 @@ enum Side { class DataArray { public: - static int END_OF_FILE; + static int NULL_WORD; static int END_OF_LINE; - static string END_OF_FILE_STR; + static string NULL_WORD_STR; static string END_OF_LINE_STR; DataArray(const string& filename); @@ -33,6 +32,8 @@ class DataArray { virtual int AtIndex(int index) const; + virtual string GetWordAtIndex(int index) const; + virtual int GetSize() const; virtual int GetVocabularySize() const; @@ -43,9 +44,12 @@ class DataArray { virtual string GetWord(int word_id) const; - int GetNumSentences() const; + virtual int GetNumSentences() const; + + virtual int GetSentenceStart(int position) const; - int GetSentenceStart(int position) const; + //TODO(pauldb): Add unit tests. + virtual int GetSentenceLength(int sentence_id) const; virtual int GetSentenceId(int position) const; diff --git a/extractor/features/count_source_target_test.cc b/extractor/features/count_source_target_test.cc new file mode 100644 index 00000000..22633bb6 --- /dev/null +++ b/extractor/features/count_source_target_test.cc @@ -0,0 +1,32 @@ +#include + +#include +#include + +#include "count_source_target.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class CountSourceTargetTest : public Test { + protected: + virtual void SetUp() { + feature = make_shared(); + } + + shared_ptr feature; +}; + +TEST_F(CountSourceTargetTest, TestGetName) { + EXPECT_EQ("CountEF", feature->GetName()); +} + +TEST_F(CountSourceTargetTest, TestScore) { + Phrase phrase; + FeatureContext context(phrase, phrase, 0.5, 9, 13); + EXPECT_EQ(1.0, feature->Score(context)); +} + +} // namespace diff --git a/extractor/features/feature.cc b/extractor/features/feature.cc index 7381c35a..876f5f8f 100644 --- a/extractor/features/feature.cc +++ b/extractor/features/feature.cc @@ -1,3 +1,5 @@ #include "feature.h" const double Feature::MAX_SCORE = 99.0; + +Feature::~Feature() {} diff --git a/extractor/features/feature.h b/extractor/features/feature.h index ad22d3e7..aca58401 100644 --- a/extractor/features/feature.h +++ b/extractor/features/feature.h @@ -10,14 +10,16 @@ using namespace std; struct FeatureContext { FeatureContext(const Phrase& source_phrase, const Phrase& target_phrase, - double sample_source_count, int pair_count) : + double source_phrase_count, int pair_count, int num_samples) : source_phrase(source_phrase), target_phrase(target_phrase), - sample_source_count(sample_source_count), pair_count(pair_count) {} + source_phrase_count(source_phrase_count), pair_count(pair_count), + num_samples(num_samples) {} Phrase source_phrase; Phrase target_phrase; - double sample_source_count; + double source_phrase_count; int pair_count; + int num_samples; }; class Feature { @@ -26,6 +28,8 @@ class Feature { virtual string GetName() const = 0; + virtual ~Feature(); + static const double MAX_SCORE; }; diff --git a/extractor/features/is_source_singleton.cc b/extractor/features/is_source_singleton.cc index 754df3bf..98d4e5fe 100644 --- a/extractor/features/is_source_singleton.cc +++ b/extractor/features/is_source_singleton.cc @@ -3,7 +3,7 @@ #include double IsSourceSingleton::Score(const FeatureContext& context) const { - return context.sample_source_count == 1; + return context.source_phrase_count == 1; } string IsSourceSingleton::GetName() const { diff --git a/extractor/features/is_source_singleton_test.cc b/extractor/features/is_source_singleton_test.cc new file mode 100644 index 00000000..8c71e593 --- /dev/null +++ b/extractor/features/is_source_singleton_test.cc @@ -0,0 +1,35 @@ +#include + +#include +#include + +#include "is_source_singleton.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class IsSourceSingletonTest : public Test { + protected: + virtual void SetUp() { + feature = make_shared(); + } + + shared_ptr feature; +}; + +TEST_F(IsSourceSingletonTest, TestGetName) { + EXPECT_EQ("IsSingletonF", feature->GetName()); +} + +TEST_F(IsSourceSingletonTest, TestScore) { + Phrase phrase; + FeatureContext context(phrase, phrase, 0.5, 3, 31); + EXPECT_EQ(0, feature->Score(context)); + + context = FeatureContext(phrase, phrase, 1, 3, 25); + EXPECT_EQ(1, feature->Score(context)); +} + +} // namespace diff --git a/extractor/features/is_source_target_singleton.cc b/extractor/features/is_source_target_singleton.cc index ec816509..31d36532 100644 --- a/extractor/features/is_source_target_singleton.cc +++ b/extractor/features/is_source_target_singleton.cc @@ -7,5 +7,5 @@ double IsSourceTargetSingleton::Score(const FeatureContext& context) const { } string IsSourceTargetSingleton::GetName() const { - return "IsSingletonEF"; + return "IsSingletonFE"; } diff --git a/extractor/features/is_source_target_singleton_test.cc b/extractor/features/is_source_target_singleton_test.cc new file mode 100644 index 00000000..a51f77c9 --- /dev/null +++ b/extractor/features/is_source_target_singleton_test.cc @@ -0,0 +1,35 @@ +#include + +#include +#include + +#include "is_source_target_singleton.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class IsSourceTargetSingletonTest : public Test { + protected: + virtual void SetUp() { + feature = make_shared(); + } + + shared_ptr feature; +}; + +TEST_F(IsSourceTargetSingletonTest, TestGetName) { + EXPECT_EQ("IsSingletonFE", feature->GetName()); +} + +TEST_F(IsSourceTargetSingletonTest, TestScore) { + Phrase phrase; + FeatureContext context(phrase, phrase, 0.5, 3, 7); + EXPECT_EQ(0, feature->Score(context)); + + context = FeatureContext(phrase, phrase, 2.3, 1, 28); + EXPECT_EQ(1, feature->Score(context)); +} + +} // namespace diff --git a/extractor/features/max_lex_source_given_target.cc b/extractor/features/max_lex_source_given_target.cc index c4792d49..21f5c76a 100644 --- a/extractor/features/max_lex_source_given_target.cc +++ b/extractor/features/max_lex_source_given_target.cc @@ -2,6 +2,7 @@ #include +#include "../data_array.h" #include "../translation_table.h" MaxLexSourceGivenTarget::MaxLexSourceGivenTarget( @@ -10,8 +11,8 @@ MaxLexSourceGivenTarget::MaxLexSourceGivenTarget( double MaxLexSourceGivenTarget::Score(const FeatureContext& context) const { vector source_words = context.source_phrase.GetWords(); - // TODO(pauldb): Add NULL to target_words, after fixing translation table. vector target_words = context.target_phrase.GetWords(); + target_words.push_back(DataArray::NULL_WORD_STR); double score = 0; for (string source_word: source_words) { @@ -26,5 +27,5 @@ double MaxLexSourceGivenTarget::Score(const FeatureContext& context) const { } string MaxLexSourceGivenTarget::GetName() const { - return "MaxLexFGivenE"; + return "MaxLexFgivenE"; } diff --git a/extractor/features/max_lex_source_given_target_test.cc b/extractor/features/max_lex_source_given_target_test.cc new file mode 100644 index 00000000..5fd41f8b --- /dev/null +++ b/extractor/features/max_lex_source_given_target_test.cc @@ -0,0 +1,74 @@ +#include + +#include +#include +#include + +#include "../mocks/mock_translation_table.h" +#include "../mocks/mock_vocabulary.h" +#include "../data_array.h" +#include "../phrase_builder.h" +#include "max_lex_source_given_target.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class MaxLexSourceGivenTargetTest : public Test { + protected: + virtual void SetUp() { + vector source_words = {"f1", "f2", "f3"}; + vector target_words = {"e1", "e2", "e3"}; + + vocabulary = make_shared(); + for (size_t i = 0; i < source_words.size(); ++i) { + EXPECT_CALL(*vocabulary, GetTerminalValue(i)) + .WillRepeatedly(Return(source_words[i])); + } + for (size_t i = 0; i < target_words.size(); ++i) { + EXPECT_CALL(*vocabulary, GetTerminalValue(i + source_words.size())) + .WillRepeatedly(Return(target_words[i])); + } + + phrase_builder = make_shared(vocabulary); + + table = make_shared(); + for (size_t i = 0; i < source_words.size(); ++i) { + for (size_t j = 0; j < target_words.size(); ++j) { + int value = i - j; + EXPECT_CALL(*table, GetSourceGivenTargetScore( + source_words[i], target_words[j])).WillRepeatedly(Return(value)); + } + } + + for (size_t i = 0; i < source_words.size(); ++i) { + int value = i * 3; + EXPECT_CALL(*table, GetSourceGivenTargetScore( + source_words[i], DataArray::NULL_WORD_STR)) + .WillRepeatedly(Return(value)); + } + + feature = make_shared(table); + } + + shared_ptr vocabulary; + shared_ptr phrase_builder; + shared_ptr table; + shared_ptr feature; +}; + +TEST_F(MaxLexSourceGivenTargetTest, TestGetName) { + EXPECT_EQ("MaxLexFgivenE", feature->GetName()); +} + +TEST_F(MaxLexSourceGivenTargetTest, TestScore) { + vector source_symbols = {0, 1, 2}; + Phrase source_phrase = phrase_builder->Build(source_symbols); + vector target_symbols = {3, 4, 5}; + Phrase target_phrase = phrase_builder->Build(target_symbols); + FeatureContext context(source_phrase, target_phrase, 0.3, 7, 11); + EXPECT_EQ(99 - log10(18), feature->Score(context)); +} + +} // namespace diff --git a/extractor/features/max_lex_target_given_source.cc b/extractor/features/max_lex_target_given_source.cc index d82182fe..f2bc2474 100644 --- a/extractor/features/max_lex_target_given_source.cc +++ b/extractor/features/max_lex_target_given_source.cc @@ -2,6 +2,7 @@ #include +#include "../data_array.h" #include "../translation_table.h" MaxLexTargetGivenSource::MaxLexTargetGivenSource( @@ -9,8 +10,8 @@ MaxLexTargetGivenSource::MaxLexTargetGivenSource( table(table) {} double MaxLexTargetGivenSource::Score(const FeatureContext& context) const { - // TODO(pauldb): Add NULL to source_words, after fixing translation table. vector source_words = context.source_phrase.GetWords(); + source_words.push_back(DataArray::NULL_WORD_STR); vector target_words = context.target_phrase.GetWords(); double score = 0; @@ -26,5 +27,5 @@ double MaxLexTargetGivenSource::Score(const FeatureContext& context) const { } string MaxLexTargetGivenSource::GetName() const { - return "MaxLexEGivenF"; + return "MaxLexEgivenF"; } diff --git a/extractor/features/max_lex_target_given_source_test.cc b/extractor/features/max_lex_target_given_source_test.cc new file mode 100644 index 00000000..c8701bf7 --- /dev/null +++ b/extractor/features/max_lex_target_given_source_test.cc @@ -0,0 +1,74 @@ +#include + +#include +#include +#include + +#include "../mocks/mock_translation_table.h" +#include "../mocks/mock_vocabulary.h" +#include "../data_array.h" +#include "../phrase_builder.h" +#include "max_lex_target_given_source.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class MaxLexTargetGivenSourceTest : public Test { + protected: + virtual void SetUp() { + vector source_words = {"f1", "f2", "f3"}; + vector target_words = {"e1", "e2", "e3"}; + + vocabulary = make_shared(); + for (size_t i = 0; i < source_words.size(); ++i) { + EXPECT_CALL(*vocabulary, GetTerminalValue(i)) + .WillRepeatedly(Return(source_words[i])); + } + for (size_t i = 0; i < target_words.size(); ++i) { + EXPECT_CALL(*vocabulary, GetTerminalValue(i + source_words.size())) + .WillRepeatedly(Return(target_words[i])); + } + + phrase_builder = make_shared(vocabulary); + + table = make_shared(); + for (size_t i = 0; i < source_words.size(); ++i) { + for (size_t j = 0; j < target_words.size(); ++j) { + int value = i - j; + EXPECT_CALL(*table, GetTargetGivenSourceScore( + source_words[i], target_words[j])).WillRepeatedly(Return(value)); + } + } + + for (size_t i = 0; i < target_words.size(); ++i) { + int value = i * 3; + EXPECT_CALL(*table, GetTargetGivenSourceScore( + DataArray::NULL_WORD_STR, target_words[i])) + .WillRepeatedly(Return(value)); + } + + feature = make_shared(table); + } + + shared_ptr vocabulary; + shared_ptr phrase_builder; + shared_ptr table; + shared_ptr feature; +}; + +TEST_F(MaxLexTargetGivenSourceTest, TestGetName) { + EXPECT_EQ("MaxLexEgivenF", feature->GetName()); +} + +TEST_F(MaxLexTargetGivenSourceTest, TestScore) { + vector source_symbols = {0, 1, 2}; + Phrase source_phrase = phrase_builder->Build(source_symbols); + vector target_symbols = {3, 4, 5}; + Phrase target_phrase = phrase_builder->Build(target_symbols); + FeatureContext context(source_phrase, target_phrase, 0.3, 7, 19); + EXPECT_EQ(-log10(36), feature->Score(context)); +} + +} // namespace diff --git a/extractor/features/sample_source_count.cc b/extractor/features/sample_source_count.cc index c8124cfb..88b645b1 100644 --- a/extractor/features/sample_source_count.cc +++ b/extractor/features/sample_source_count.cc @@ -3,7 +3,7 @@ #include double SampleSourceCount::Score(const FeatureContext& context) const { - return log10(1 + context.sample_source_count); + return log10(1 + context.num_samples); } string SampleSourceCount::GetName() const { diff --git a/extractor/features/sample_source_count_test.cc b/extractor/features/sample_source_count_test.cc new file mode 100644 index 00000000..7d226104 --- /dev/null +++ b/extractor/features/sample_source_count_test.cc @@ -0,0 +1,36 @@ +#include + +#include +#include +#include + +#include "sample_source_count.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class SampleSourceCountTest : public Test { + protected: + virtual void SetUp() { + feature = make_shared(); + } + + shared_ptr feature; +}; + +TEST_F(SampleSourceCountTest, TestGetName) { + EXPECT_EQ("SampleCountF", feature->GetName()); +} + +TEST_F(SampleSourceCountTest, TestScore) { + Phrase phrase; + FeatureContext context(phrase, phrase, 0, 3, 1); + EXPECT_EQ(log10(2), feature->Score(context)); + + context = FeatureContext(phrase, phrase, 3.2, 3, 9); + EXPECT_EQ(1.0, feature->Score(context)); +} + +} // namespace diff --git a/extractor/features/target_given_source_coherent.cc b/extractor/features/target_given_source_coherent.cc index 748413c3..274b3364 100644 --- a/extractor/features/target_given_source_coherent.cc +++ b/extractor/features/target_given_source_coherent.cc @@ -3,10 +3,10 @@ #include double TargetGivenSourceCoherent::Score(const FeatureContext& context) const { - double prob = context.pair_count / context.sample_source_count; + double prob = (double) context.pair_count / context.num_samples; return prob > 0 ? -log10(prob) : MAX_SCORE; } string TargetGivenSourceCoherent::GetName() const { - return "EGivenFCoherent"; + return "EgivenFCoherent"; } diff --git a/extractor/features/target_given_source_coherent_test.cc b/extractor/features/target_given_source_coherent_test.cc new file mode 100644 index 00000000..c54c06c2 --- /dev/null +++ b/extractor/features/target_given_source_coherent_test.cc @@ -0,0 +1,35 @@ +#include + +#include +#include + +#include "target_given_source_coherent.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class TargetGivenSourceCoherentTest : public Test { + protected: + virtual void SetUp() { + feature = make_shared(); + } + + shared_ptr feature; +}; + +TEST_F(TargetGivenSourceCoherentTest, TestGetName) { + EXPECT_EQ("EgivenFCoherent", feature->GetName()); +} + +TEST_F(TargetGivenSourceCoherentTest, TestScore) { + Phrase phrase; + FeatureContext context(phrase, phrase, 0.3, 2, 20); + EXPECT_EQ(1.0, feature->Score(context)); + + context = FeatureContext(phrase, phrase, 1.9, 0, 1); + EXPECT_EQ(99.0, feature->Score(context)); +} + +} // namespace diff --git a/extractor/grammar.cc b/extractor/grammar.cc index 79a0541d..8124a804 100644 --- a/extractor/grammar.cc +++ b/extractor/grammar.cc @@ -1,17 +1,32 @@ #include "grammar.h" +#include + #include "rule.h" +using namespace std; + Grammar::Grammar(const vector& rules, const vector& feature_names) : rules(rules), feature_names(feature_names) {} +vector Grammar::GetRules() const { + return rules; +} + +vector Grammar::GetFeatureNames() const { + return feature_names; +} + ostream& operator<<(ostream& os, const Grammar& grammar) { - for (Rule rule: grammar.rules) { + vector rules = grammar.GetRules(); + vector feature_names = grammar.GetFeatureNames(); + os << setprecision(12); + for (Rule rule: rules) { os << "[X] ||| " << rule.source_phrase << " ||| " << rule.target_phrase << " |||"; for (size_t i = 0; i < rule.scores.size(); ++i) { - os << " " << grammar.feature_names[i] << "=" << rule.scores[i]; + os << " " << feature_names[i] << "=" << rule.scores[i]; } os << " |||"; for (auto link: rule.alignment) { diff --git a/extractor/grammar.h b/extractor/grammar.h index db15fa7e..889cc2f3 100644 --- a/extractor/grammar.h +++ b/extractor/grammar.h @@ -13,6 +13,10 @@ class Grammar { public: Grammar(const vector& rules, const vector& feature_names); + vector GetRules() const; + + vector GetFeatureNames() const; + friend ostream& operator<<(ostream& os, const Grammar& grammar); private: diff --git a/extractor/grammar_extractor.cc b/extractor/grammar_extractor.cc index 15268165..2f008026 100644 --- a/extractor/grammar_extractor.cc +++ b/extractor/grammar_extractor.cc @@ -10,19 +10,6 @@ using namespace std; -vector Tokenize(const string& sentence) { - vector result; - result.push_back(""); - - istringstream buffer(sentence); - copy(istream_iterator(buffer), - istream_iterator(), - back_inserter(result)); - - result.push_back(""); - return result; -} - GrammarExtractor::GrammarExtractor( shared_ptr source_suffix_array, shared_ptr target_data_array, @@ -31,15 +18,35 @@ GrammarExtractor::GrammarExtractor( int max_nonterminals, int max_rule_symbols, int max_samples, bool use_baeza_yates, bool require_tight_phrases) : vocabulary(make_shared()), - rule_factory(source_suffix_array, target_data_array, alignment, - vocabulary, precomputation, scorer, min_gap_size, max_rule_span, - max_nonterminals, max_rule_symbols, max_samples, use_baeza_yates, - require_tight_phrases) {} + rule_factory(make_shared( + source_suffix_array, target_data_array, alignment, vocabulary, + precomputation, scorer, min_gap_size, max_rule_span, max_nonterminals, + max_rule_symbols, max_samples, use_baeza_yates, + require_tight_phrases)) {} + +GrammarExtractor::GrammarExtractor( + shared_ptr vocabulary, + shared_ptr rule_factory) : + vocabulary(vocabulary), + rule_factory(rule_factory) {} Grammar GrammarExtractor::GetGrammar(const string& sentence) { - vector words = Tokenize(sentence); + vector words = TokenizeSentence(sentence); vector word_ids = AnnotateWords(words); - return rule_factory.GetGrammar(word_ids); + return rule_factory->GetGrammar(word_ids); +} + +vector GrammarExtractor::TokenizeSentence(const string& sentence) { + vector result; + result.push_back(""); + + istringstream buffer(sentence); + copy(istream_iterator(buffer), + istream_iterator(), + back_inserter(result)); + + result.push_back(""); + return result; } vector GrammarExtractor::AnnotateWords(const vector& words) { diff --git a/extractor/grammar_extractor.h b/extractor/grammar_extractor.h index 243f33cf..5f87faa7 100644 --- a/extractor/grammar_extractor.h +++ b/extractor/grammar_extractor.h @@ -32,13 +32,19 @@ class GrammarExtractor { bool use_baeza_yates, bool require_tight_phrases); + // For testing only. + GrammarExtractor(shared_ptr vocabulary, + shared_ptr rule_factory); + Grammar GetGrammar(const string& sentence); private: + vector TokenizeSentence(const string& sentence); + vector AnnotateWords(const vector& words); shared_ptr vocabulary; - HieroCachingRuleFactory rule_factory; + shared_ptr rule_factory; }; #endif diff --git a/extractor/grammar_extractor_test.cc b/extractor/grammar_extractor_test.cc new file mode 100644 index 00000000..d4ed7d4f --- /dev/null +++ b/extractor/grammar_extractor_test.cc @@ -0,0 +1,49 @@ +#include + +#include +#include +#include + +#include "grammar.h" +#include "grammar_extractor.h" +#include "mocks/mock_rule_factory.h" +#include "mocks/mock_vocabulary.h" +#include "rule.h" + +using namespace std; +using namespace ::testing; + +namespace { + +TEST(GrammarExtractorTest, TestAnnotatingWords) { + shared_ptr vocabulary = make_shared(); + EXPECT_CALL(*vocabulary, GetTerminalIndex("")) + .WillRepeatedly(Return(0)); + EXPECT_CALL(*vocabulary, GetTerminalIndex("Anna")) + .WillRepeatedly(Return(1)); + EXPECT_CALL(*vocabulary, GetTerminalIndex("has")) + .WillRepeatedly(Return(2)); + EXPECT_CALL(*vocabulary, GetTerminalIndex("many")) + .WillRepeatedly(Return(3)); + EXPECT_CALL(*vocabulary, GetTerminalIndex("apples")) + .WillRepeatedly(Return(4)); + EXPECT_CALL(*vocabulary, GetTerminalIndex(".")) + .WillRepeatedly(Return(5)); + EXPECT_CALL(*vocabulary, GetTerminalIndex("")) + .WillRepeatedly(Return(6)); + + shared_ptr factory = + make_shared(); + vector word_ids = {0, 1, 2, 3, 3, 4, 5, 6}; + vector rules; + vector feature_names; + Grammar grammar(rules, feature_names); + EXPECT_CALL(*factory, GetGrammar(word_ids)) + .WillOnce(Return(grammar)); + + GrammarExtractor extractor(vocabulary, factory); + string sentence = "Anna has many many apples ."; + extractor.GetGrammar(sentence); +} + +} // namespace diff --git a/extractor/intersector.cc b/extractor/intersector.cc index b53479af..cf42f630 100644 --- a/extractor/intersector.cc +++ b/extractor/intersector.cc @@ -1,5 +1,7 @@ #include "intersector.h" +#include + #include "data_array.h" #include "matching_comparator.h" #include "phrase.h" @@ -9,6 +11,10 @@ #include "veb.h" #include "vocabulary.h" +using namespace std::chrono; + +typedef high_resolution_clock Clock; + Intersector::Intersector(shared_ptr vocabulary, shared_ptr precomputation, shared_ptr suffix_array, @@ -38,12 +44,22 @@ Intersector::Intersector(shared_ptr vocabulary, ConvertIndexes(precomputation, suffix_array->GetData()); } +Intersector::Intersector() {} + +Intersector::~Intersector() {} + void Intersector::ConvertIndexes(shared_ptr precomputation, shared_ptr data_array) { const Index& precomputed_index = precomputation->GetInvertedIndex(); for (pair, vector > entry: precomputed_index) { vector phrase = ConvertPhrase(entry.first, data_array); inverted_index[phrase] = entry.second; + + phrase.push_back(vocabulary->GetNonterminalIndex(1)); + inverted_index[phrase] = entry.second; + phrase.pop_back(); + phrase.insert(phrase.begin(), vocabulary->GetNonterminalIndex(1)); + inverted_index[phrase] = entry.second; } const Index& precomputed_collocations = precomputation->GetCollocations(); @@ -76,6 +92,9 @@ PhraseLocation Intersector::Intersect( const Phrase& prefix, PhraseLocation& prefix_location, const Phrase& suffix, PhraseLocation& suffix_location, const Phrase& phrase) { + if (linear_merge_time == 0) { + linear_merger->linear_merge_time = 0; + } vector symbols = phrase.Get(); // We should never attempt to do an intersect query for a pattern starting or @@ -95,17 +114,23 @@ PhraseLocation Intersector::Intersect( shared_ptr > prefix_matchings = prefix_location.matchings; shared_ptr > suffix_matchings = suffix_location.matchings; int prefix_subpatterns = prefix_location.num_subpatterns; - int suffix_subpatterns = prefix_location.num_subpatterns; + int suffix_subpatterns = suffix_location.num_subpatterns; if (use_baeza_yates) { + double prev_linear_merge_time = linear_merger->linear_merge_time; + Clock::time_point start = Clock::now(); binary_search_merger->Merge(locations, phrase, suffix, prefix_matchings->begin(), prefix_matchings->end(), suffix_matchings->begin(), suffix_matchings->end(), prefix_subpatterns, suffix_subpatterns); + Clock::time_point stop = Clock::now(); + binary_merge_time += duration_cast(stop - start).count() - + (linear_merger->linear_merge_time - prev_linear_merge_time); } else { linear_merger->Merge(locations, phrase, suffix, prefix_matchings->begin(), prefix_matchings->end(), suffix_matchings->begin(), suffix_matchings->end(), prefix_subpatterns, suffix_subpatterns); } + linear_merge_time = linear_merger->linear_merge_time; return PhraseLocation(locations, phrase.Arity() + 1); } @@ -116,6 +141,8 @@ void Intersector::ExtendPhraseLocation( return; } + Clock::time_point sort_start = Clock::now(); + phrase_location.num_subpatterns = 1; phrase_location.sa_low = phrase_location.sa_high = 0; @@ -140,4 +167,6 @@ void Intersector::ExtendPhraseLocation( } phrase_location.matchings = make_shared >(matchings); + Clock::time_point sort_stop = Clock::now(); + sort_time += duration_cast(sort_stop - sort_start).count(); } diff --git a/extractor/intersector.h b/extractor/intersector.h index f023cc96..8b159f17 100644 --- a/extractor/intersector.h +++ b/extractor/intersector.h @@ -2,7 +2,7 @@ #define _INTERSECTOR_H_ #include -#include +#include #include #include @@ -11,7 +11,6 @@ #include "linear_merger.h" using namespace std; -using namespace tr1; typedef boost::hash > VectorHash; typedef unordered_map, vector, VectorHash> Index; @@ -42,11 +41,16 @@ class Intersector { shared_ptr binary_search_merger, bool use_baeza_yates); - PhraseLocation Intersect( + virtual ~Intersector(); + + virtual PhraseLocation Intersect( const Phrase& prefix, PhraseLocation& prefix_location, const Phrase& suffix, PhraseLocation& suffix_location, const Phrase& phrase); + protected: + Intersector(); + private: void ConvertIndexes(shared_ptr precomputation, shared_ptr data_array); @@ -64,6 +68,12 @@ class Intersector { Index inverted_index; Index collocations; bool use_baeza_yates; + + // TODO(pauldb): Don't forget to remove these. + public: + double sort_time; + double linear_merge_time; + double binary_merge_time; }; #endif diff --git a/extractor/intersector_test.cc b/extractor/intersector_test.cc index a3756902..ec318362 100644 --- a/extractor/intersector_test.cc +++ b/extractor/intersector_test.cc @@ -34,7 +34,7 @@ class IntersectorTest : public Test { .WillRepeatedly(Return(words[i])); } - vector suffixes = {0, 1, 3, 5, 2, 4, 6}; + vector suffixes = {6, 0, 5, 3, 1, 4, 2}; suffix_array = make_shared(); EXPECT_CALL(*suffix_array, GetData()) .WillRepeatedly(Return(data_array)); @@ -103,7 +103,7 @@ TEST_F(IntersectorTest, TestLinearMergeaXb) { Phrase suffix = phrase_builder->Build(suffix_symbols); vector symbols = {3, -1, 4}; Phrase phrase = phrase_builder->Build(symbols); - PhraseLocation prefix_locs(1, 4), suffix_locs(4, 6); + PhraseLocation prefix_locs(2, 5), suffix_locs(5, 7); vector ex_prefix_locs = {1, 3, 5}; PhraseLocation extended_prefix_locs(ex_prefix_locs, 1); @@ -135,7 +135,7 @@ TEST_F(IntersectorTest, TestBinarySearchMergeaXb) { Phrase suffix = phrase_builder->Build(suffix_symbols); vector symbols = {3, -1, 4}; Phrase phrase = phrase_builder->Build(symbols); - PhraseLocation prefix_locs(1, 4), suffix_locs(4, 6); + PhraseLocation prefix_locs(2, 5), suffix_locs(5, 7); vector ex_prefix_locs = {1, 3, 5}; PhraseLocation extended_prefix_locs(ex_prefix_locs, 1); diff --git a/extractor/linear_merger.cc b/extractor/linear_merger.cc index 666f8d87..7233f945 100644 --- a/extractor/linear_merger.cc +++ b/extractor/linear_merger.cc @@ -1,5 +1,6 @@ #include "linear_merger.h" +#include #include #include "data_array.h" @@ -9,6 +10,10 @@ #include "phrase_location.h" #include "vocabulary.h" +using namespace std::chrono; + +typedef high_resolution_clock Clock; + LinearMerger::LinearMerger(shared_ptr vocabulary, shared_ptr data_array, shared_ptr comparator) : @@ -22,7 +27,9 @@ void LinearMerger::Merge( vector& locations, const Phrase& phrase, const Phrase& suffix, vector::iterator prefix_start, vector::iterator prefix_end, vector::iterator suffix_start, vector::iterator suffix_end, - int prefix_subpatterns, int suffix_subpatterns) const { + int prefix_subpatterns, int suffix_subpatterns) { + Clock::time_point start = Clock::now(); + int last_chunk_len = suffix.GetChunkLen(suffix.Arity()); bool offset = !vocabulary->IsTerminal(suffix.GetSymbol(0)); @@ -62,4 +69,7 @@ void LinearMerger::Merge( prefix_start += prefix_subpatterns; } } + + Clock::time_point stop = Clock::now(); + linear_merge_time += duration_cast(stop - start).count(); } diff --git a/extractor/linear_merger.h b/extractor/linear_merger.h index 6a69b804..25692b15 100644 --- a/extractor/linear_merger.h +++ b/extractor/linear_merger.h @@ -24,7 +24,7 @@ class LinearMerger { vector& locations, const Phrase& phrase, const Phrase& suffix, vector::iterator prefix_start, vector::iterator prefix_end, vector::iterator suffix_start, vector::iterator suffix_end, - int prefix_subpatterns, int suffix_subpatterns) const; + int prefix_subpatterns, int suffix_subpatterns); protected: LinearMerger(); @@ -33,6 +33,10 @@ class LinearMerger { shared_ptr vocabulary; shared_ptr data_array; shared_ptr comparator; + + // TODO(pauldb): Remove this eventually. + public: + double linear_merge_time; }; #endif diff --git a/extractor/matchings_finder.cc b/extractor/matchings_finder.cc index ba4edab1..eaf493b2 100644 --- a/extractor/matchings_finder.cc +++ b/extractor/matchings_finder.cc @@ -6,6 +6,10 @@ MatchingsFinder::MatchingsFinder(shared_ptr suffix_array) : suffix_array(suffix_array) {} +MatchingsFinder::MatchingsFinder() {} + +MatchingsFinder::~MatchingsFinder() {} + PhraseLocation MatchingsFinder::Find(PhraseLocation& location, const string& word, int offset) { if (location.sa_low == -1 && location.sa_high == -1) { diff --git a/extractor/matchings_finder.h b/extractor/matchings_finder.h index 0458a4d8..ed04d8b8 100644 --- a/extractor/matchings_finder.h +++ b/extractor/matchings_finder.h @@ -13,7 +13,13 @@ class MatchingsFinder { public: MatchingsFinder(shared_ptr suffix_array); - PhraseLocation Find(PhraseLocation& location, const string& word, int offset); + virtual ~MatchingsFinder(); + + virtual PhraseLocation Find(PhraseLocation& location, const string& word, + int offset); + + protected: + MatchingsFinder(); private: shared_ptr suffix_array; diff --git a/extractor/matchings_trie.cc b/extractor/matchings_trie.cc index 851d4596..921ec582 100644 --- a/extractor/matchings_trie.cc +++ b/extractor/matchings_trie.cc @@ -1,11 +1,19 @@ #include "matchings_trie.h" void MatchingsTrie::Reset() { - // TODO(pauldb): This is probably memory leaking because of the suffix links. - // Check if it's true and free the memory properly. - root.reset(new TrieNode()); + ResetTree(root); + root = make_shared(); } shared_ptr MatchingsTrie::GetRoot() const { return root; } + +void MatchingsTrie::ResetTree(shared_ptr root) { + if (root != NULL) { + for (auto child: root->children) { + ResetTree(child.second); + } + root.reset(); + } +} diff --git a/extractor/matchings_trie.h b/extractor/matchings_trie.h index f935d1a9..6e72b2db 100644 --- a/extractor/matchings_trie.h +++ b/extractor/matchings_trie.h @@ -2,13 +2,12 @@ #define _MATCHINGS_TRIE_ #include -#include +#include #include "phrase.h" #include "phrase_location.h" using namespace std; -using namespace tr1; struct TrieNode { TrieNode(shared_ptr suffix_link = shared_ptr(), @@ -40,6 +39,8 @@ class MatchingsTrie { shared_ptr GetRoot() const; private: + void ResetTree(shared_ptr root); + shared_ptr root; }; diff --git a/extractor/mocks/mock_alignment.h b/extractor/mocks/mock_alignment.h new file mode 100644 index 00000000..4a5077ad --- /dev/null +++ b/extractor/mocks/mock_alignment.h @@ -0,0 +1,10 @@ +#include + +#include "../alignment.h" + +typedef vector > SentenceLinks; + +class MockAlignment : public Alignment { + public: + MOCK_CONST_METHOD1(GetLinks, SentenceLinks(int sentence_id)); +}; diff --git a/extractor/mocks/mock_binary_search_merger.h b/extractor/mocks/mock_binary_search_merger.h index e1375ee3..e23386f0 100644 --- a/extractor/mocks/mock_binary_search_merger.h +++ b/extractor/mocks/mock_binary_search_merger.h @@ -10,6 +10,6 @@ using namespace std; class MockBinarySearchMerger: public BinarySearchMerger { public: MOCK_CONST_METHOD9(Merge, void(vector&, const Phrase&, const Phrase&, - vector::iterator, vector::iterator, vector::iterator, - vector::iterator, int, int)); + const vector::iterator&, const vector::iterator&, + const vector::iterator&, const vector::iterator&, int, int)); }; diff --git a/extractor/mocks/mock_data_array.h b/extractor/mocks/mock_data_array.h index 54497cf5..004e8906 100644 --- a/extractor/mocks/mock_data_array.h +++ b/extractor/mocks/mock_data_array.h @@ -6,10 +6,14 @@ class MockDataArray : public DataArray { public: MOCK_CONST_METHOD0(GetData, const vector&()); MOCK_CONST_METHOD1(AtIndex, int(int index)); + MOCK_CONST_METHOD1(GetWordAtIndex, string(int index)); MOCK_CONST_METHOD0(GetSize, int()); MOCK_CONST_METHOD0(GetVocabularySize, int()); MOCK_CONST_METHOD1(HasWord, bool(const string& word)); MOCK_CONST_METHOD1(GetWordId, int(const string& word)); MOCK_CONST_METHOD1(GetWord, string(int word_id)); + MOCK_CONST_METHOD1(GetSentenceLength, int(int sentence_id)); + MOCK_CONST_METHOD0(GetNumSentences, int()); + MOCK_CONST_METHOD1(GetSentenceStart, int(int sentence_id)); MOCK_CONST_METHOD1(GetSentenceId, int(int position)); }; diff --git a/extractor/mocks/mock_feature.h b/extractor/mocks/mock_feature.h new file mode 100644 index 00000000..d2137629 --- /dev/null +++ b/extractor/mocks/mock_feature.h @@ -0,0 +1,9 @@ +#include + +#include "../features/feature.h" + +class MockFeature : public Feature { + public: + MOCK_CONST_METHOD1(Score, double(const FeatureContext& context)); + MOCK_CONST_METHOD0(GetName, string()); +}; diff --git a/extractor/mocks/mock_intersector.h b/extractor/mocks/mock_intersector.h new file mode 100644 index 00000000..372fa7ea --- /dev/null +++ b/extractor/mocks/mock_intersector.h @@ -0,0 +1,11 @@ +#include + +#include "../intersector.h" +#include "../phrase.h" +#include "../phrase_location.h" + +class MockIntersector : public Intersector { + public: + MOCK_METHOD5(Intersect, PhraseLocation(const Phrase&, PhraseLocation&, + const Phrase&, PhraseLocation&, const Phrase&)); +}; diff --git a/extractor/mocks/mock_linear_merger.h b/extractor/mocks/mock_linear_merger.h index 82243428..522c1f31 100644 --- a/extractor/mocks/mock_linear_merger.h +++ b/extractor/mocks/mock_linear_merger.h @@ -9,7 +9,7 @@ using namespace std; class MockLinearMerger: public LinearMerger { public: - MOCK_CONST_METHOD9(Merge, void(vector&, const Phrase&, const Phrase&, + MOCK_METHOD9(Merge, void(vector&, const Phrase&, const Phrase&, vector::iterator, vector::iterator, vector::iterator, vector::iterator, int, int)); }; diff --git a/extractor/mocks/mock_matchings_finder.h b/extractor/mocks/mock_matchings_finder.h new file mode 100644 index 00000000..3e80d266 --- /dev/null +++ b/extractor/mocks/mock_matchings_finder.h @@ -0,0 +1,9 @@ +#include + +#include "../matchings_finder.h" +#include "../phrase_location.h" + +class MockMatchingsFinder : public MatchingsFinder { + public: + MOCK_METHOD3(Find, PhraseLocation(PhraseLocation&, const string&, int)); +}; diff --git a/extractor/mocks/mock_rule_extractor.h b/extractor/mocks/mock_rule_extractor.h new file mode 100644 index 00000000..f18e009a --- /dev/null +++ b/extractor/mocks/mock_rule_extractor.h @@ -0,0 +1,12 @@ +#include + +#include "../phrase.h" +#include "../phrase_builder.h" +#include "../rule.h" +#include "../rule_extractor.h" + +class MockRuleExtractor : public RuleExtractor { + public: + MOCK_CONST_METHOD2(ExtractRules, vector(const Phrase&, + const PhraseLocation&)); +}; diff --git a/extractor/mocks/mock_rule_extractor_helper.h b/extractor/mocks/mock_rule_extractor_helper.h new file mode 100644 index 00000000..63ff1048 --- /dev/null +++ b/extractor/mocks/mock_rule_extractor_helper.h @@ -0,0 +1,78 @@ +#include + +#include + +#include "../rule_extractor_helper.h" + +using namespace std; + +typedef unordered_map Indexes; + +class MockRuleExtractorHelper : public RuleExtractorHelper { + public: + MOCK_CONST_METHOD5(GetLinksSpans, void(vector&, vector&, + vector&, vector&, int)); + MOCK_CONST_METHOD3(CheckAlignedTerminals, bool(const vector&, + const vector&, const vector&)); + MOCK_CONST_METHOD3(CheckTightPhrases, bool(const vector&, + const vector&, const vector&)); + MOCK_CONST_METHOD1(GetGapOrder, vector(const vector >&)); + MOCK_CONST_METHOD3(GetSourceIndexes, Indexes(const vector&, + const vector&, int)); + + // We need to implement these methods, because Google Mock doesn't support + // methods with more than 10 arguments. + bool FindFixPoint( + int, int, const vector&, const vector&, int& target_phrase_low, + int& target_phrase_high, const vector&, const vector&, + int& source_back_low, int& source_back_high, int, int, int, int, bool, + bool, bool) const { + target_phrase_low = this->target_phrase_low; + target_phrase_high = this->target_phrase_high; + source_back_low = this->source_back_low; + source_back_high = this->source_back_high; + return find_fix_point; + } + + bool GetGaps(vector >& source_gaps, + vector >& target_gaps, + const vector&, const vector&, const vector&, + const vector&, const vector&, const vector&, + int, int, int, int, int& num_symbols, + bool& met_constraints) const { + source_gaps = this->source_gaps; + target_gaps = this->target_gaps; + num_symbols = this->num_symbols; + met_constraints = this->met_constraints; + return get_gaps; + } + + void SetUp( + int target_phrase_low, int target_phrase_high, int source_back_low, + int source_back_high, bool find_fix_point, + vector > source_gaps, vector > target_gaps, + int num_symbols, bool met_constraints, bool get_gaps) { + this->target_phrase_low = target_phrase_low; + this->target_phrase_high = target_phrase_high; + this->source_back_low = source_back_low; + this->source_back_high = source_back_high; + this->find_fix_point = find_fix_point; + this->source_gaps = source_gaps; + this->target_gaps = target_gaps; + this->num_symbols = num_symbols; + this->met_constraints = met_constraints; + this->get_gaps = get_gaps; + } + + private: + int target_phrase_low; + int target_phrase_high; + int source_back_low; + int source_back_high; + bool find_fix_point; + vector > source_gaps; + vector > target_gaps; + int num_symbols; + bool met_constraints; + bool get_gaps; +}; diff --git a/extractor/mocks/mock_rule_factory.h b/extractor/mocks/mock_rule_factory.h new file mode 100644 index 00000000..2a96be93 --- /dev/null +++ b/extractor/mocks/mock_rule_factory.h @@ -0,0 +1,9 @@ +#include + +#include "../grammar.h" +#include "../rule_factory.h" + +class MockHieroCachingRuleFactory : public HieroCachingRuleFactory { + public: + MOCK_METHOD1(GetGrammar, Grammar(const vector& word_ids)); +}; diff --git a/extractor/mocks/mock_sampler.h b/extractor/mocks/mock_sampler.h new file mode 100644 index 00000000..b2306109 --- /dev/null +++ b/extractor/mocks/mock_sampler.h @@ -0,0 +1,9 @@ +#include + +#include "../phrase_location.h" +#include "../sampler.h" + +class MockSampler : public Sampler { + public: + MOCK_CONST_METHOD1(Sample, PhraseLocation(const PhraseLocation& location)); +}; diff --git a/extractor/mocks/mock_scorer.h b/extractor/mocks/mock_scorer.h new file mode 100644 index 00000000..48115ef4 --- /dev/null +++ b/extractor/mocks/mock_scorer.h @@ -0,0 +1,10 @@ +#include + +#include "../scorer.h" +#include "../features/feature.h" + +class MockScorer : public Scorer { + public: + MOCK_CONST_METHOD1(Score, vector(const FeatureContext& context)); + MOCK_CONST_METHOD0(GetFeatureNames, vector()); +}; diff --git a/extractor/mocks/mock_target_phrase_extractor.h b/extractor/mocks/mock_target_phrase_extractor.h new file mode 100644 index 00000000..6dc6bba6 --- /dev/null +++ b/extractor/mocks/mock_target_phrase_extractor.h @@ -0,0 +1,12 @@ +#include + +#include "../target_phrase_extractor.h" + +typedef pair PhraseExtract; + +class MockTargetPhraseExtractor : public TargetPhraseExtractor { + public: + MOCK_CONST_METHOD6(ExtractPhrases, vector( + const vector > &, const vector&, int, int, + const unordered_map&, int)); +}; diff --git a/extractor/mocks/mock_translation_table.h b/extractor/mocks/mock_translation_table.h new file mode 100644 index 00000000..a35c9327 --- /dev/null +++ b/extractor/mocks/mock_translation_table.h @@ -0,0 +1,9 @@ +#include + +#include "../translation_table.h" + +class MockTranslationTable : public TranslationTable { + public: + MOCK_METHOD2(GetSourceGivenTargetScore, double(const string&, const string&)); + MOCK_METHOD2(GetTargetGivenSourceScore, double(const string&, const string&)); +}; diff --git a/extractor/phrase_builder.cc b/extractor/phrase_builder.cc index c4e0c2ed..4325390c 100644 --- a/extractor/phrase_builder.cc +++ b/extractor/phrase_builder.cc @@ -9,10 +9,9 @@ PhraseBuilder::PhraseBuilder(shared_ptr vocabulary) : Phrase PhraseBuilder::Build(const vector& symbols) { Phrase phrase; phrase.symbols = symbols; - phrase.words.resize(symbols.size()); for (size_t i = 0; i < symbols.size(); ++i) { if (vocabulary->IsTerminal(symbols[i])) { - phrase.words[i] = vocabulary->GetTerminalValue(symbols[i]); + phrase.words.push_back(vocabulary->GetTerminalValue(symbols[i])); } else { phrase.var_pos.push_back(i); } @@ -30,7 +29,7 @@ Phrase PhraseBuilder::Extend(const Phrase& phrase, bool start_x, bool end_x) { } for (size_t i = start_x; i < symbols.size(); ++i) { - if (vocabulary->IsTerminal(symbols[i])) { + if (!vocabulary->IsTerminal(symbols[i])) { ++num_nonterminals; symbols[i] = vocabulary->GetNonterminalIndex(num_nonterminals); } diff --git a/extractor/phrase_location.cc b/extractor/phrase_location.cc index 984407c5..62f1e714 100644 --- a/extractor/phrase_location.cc +++ b/extractor/phrase_location.cc @@ -10,7 +10,11 @@ PhraseLocation::PhraseLocation(const vector& matchings, num_subpatterns(num_subpatterns) {} bool PhraseLocation::IsEmpty() { - return sa_low >= sa_high || (num_subpatterns > 0 && matchings->size() == 0); + if (num_subpatterns > 0) { + return matchings->size() == 0; + } else { + return sa_low >= sa_high; + } } bool operator==(const PhraseLocation& a, const PhraseLocation& b) { diff --git a/extractor/precomputation.cc b/extractor/precomputation.cc index 9a167976..8a76beb1 100644 --- a/extractor/precomputation.cc +++ b/extractor/precomputation.cc @@ -7,7 +7,6 @@ #include "suffix_array.h" using namespace std; -using namespace tr1; int Precomputation::NON_TERMINAL = -1; @@ -79,13 +78,16 @@ vector > Precomputation::FindMostFrequentPatterns( } vector > frequent_patterns; - for (size_t i = 0; i < num_frequent_patterns && !heap.empty(); ++i) { + while (frequent_patterns.size() < num_frequent_patterns && !heap.empty()) { int start = heap.top().second.first; int len = heap.top().second.second; heap.pop(); vector pattern(data.begin() + start, data.begin() + start + len); - frequent_patterns.push_back(pattern); + if (find(pattern.begin(), pattern.end(), DataArray::END_OF_LINE) == + pattern.end()) { + frequent_patterns.push_back(pattern); + } } return frequent_patterns; } diff --git a/extractor/precomputation.h b/extractor/precomputation.h index 428505d8..28426bfa 100644 --- a/extractor/precomputation.h +++ b/extractor/precomputation.h @@ -2,8 +2,8 @@ #define _PRECOMPUTATION_H_ #include -#include -#include +#include +#include #include #include @@ -12,7 +12,6 @@ namespace fs = boost::filesystem; using namespace std; -using namespace tr1; class SuffixArray; diff --git a/extractor/rule_extractor.cc b/extractor/rule_extractor.cc index 9460020f..92343241 100644 --- a/extractor/rule_extractor.cc +++ b/extractor/rule_extractor.cc @@ -1,7 +1,6 @@ #include "rule_extractor.h" #include -#include #include "alignment.h" #include "data_array.h" @@ -9,11 +8,11 @@ #include "phrase_builder.h" #include "phrase_location.h" #include "rule.h" +#include "rule_extractor_helper.h" #include "scorer.h" -#include "vocabulary.h" +#include "target_phrase_extractor.h" using namespace std; -using namespace tr1; RuleExtractor::RuleExtractor( shared_ptr source_data_array, @@ -29,20 +28,50 @@ RuleExtractor::RuleExtractor( bool require_aligned_terminal, bool require_aligned_chunks, bool require_tight_phrases) : - source_data_array(source_data_array), target_data_array(target_data_array), - alignment(alignment), + source_data_array(source_data_array), phrase_builder(phrase_builder), scorer(scorer), - vocabulary(vocabulary), max_rule_span(max_rule_span), min_gap_size(min_gap_size), max_nonterminals(max_nonterminals), max_rule_symbols(max_rule_symbols), - require_aligned_terminal(require_aligned_terminal), - require_aligned_chunks(require_aligned_chunks), + require_tight_phrases(require_tight_phrases) { + helper = make_shared( + source_data_array, target_data_array, alignment, max_rule_span, + max_rule_symbols, require_aligned_terminal, require_aligned_chunks, + require_tight_phrases); + target_phrase_extractor = make_shared( + target_data_array, alignment, phrase_builder, helper, vocabulary, + max_rule_span, require_tight_phrases); +} + +RuleExtractor::RuleExtractor( + shared_ptr source_data_array, + shared_ptr phrase_builder, + shared_ptr scorer, + shared_ptr target_phrase_extractor, + shared_ptr helper, + int max_rule_span, + int min_gap_size, + int max_nonterminals, + int max_rule_symbols, + bool require_tight_phrases) : + source_data_array(source_data_array), + phrase_builder(phrase_builder), + scorer(scorer), + target_phrase_extractor(target_phrase_extractor), + helper(helper), + max_rule_span(max_rule_span), + min_gap_size(min_gap_size), + max_nonterminals(max_nonterminals), + max_rule_symbols(max_rule_symbols), require_tight_phrases(require_tight_phrases) {} +RuleExtractor::RuleExtractor() {} + +RuleExtractor::~RuleExtractor() {} + vector RuleExtractor::ExtractRules(const Phrase& phrase, const PhraseLocation& location) const { int num_subpatterns = location.num_subpatterns; @@ -60,6 +89,7 @@ vector RuleExtractor::ExtractRules(const Phrase& phrase, } } + int num_samples = matchings.size() / num_subpatterns; vector rules; for (auto source_phrase_entry: alignments_counter) { Phrase source_phrase = source_phrase_entry.first; @@ -77,7 +107,7 @@ vector RuleExtractor::ExtractRules(const Phrase& phrase, } FeatureContext context(source_phrase, target_phrase, - source_phrase_counter[source_phrase], num_locations); + source_phrase_counter[source_phrase], num_locations, num_samples); vector scores = scorer->Score(context); rules.push_back(Rule(source_phrase, target_phrase, scores, most_frequent_alignment)); @@ -93,7 +123,8 @@ vector RuleExtractor::ExtractAlignments( int source_sent_start = source_data_array->GetSentenceStart(sentence_id); vector source_low, source_high, target_low, target_high; - GetLinksSpans(source_low, source_high, target_low, target_high, sentence_id); + helper->GetLinksSpans(source_low, source_high, target_low, target_high, + sentence_id); int num_subpatterns = matching.size(); vector chunklen(num_subpatterns); @@ -101,39 +132,44 @@ vector RuleExtractor::ExtractAlignments( chunklen[i] = phrase.GetChunkLen(i); } - if (!CheckAlignedTerminals(matching, chunklen, source_low) || - !CheckTightPhrases(matching, chunklen, source_low)) { + if (!helper->CheckAlignedTerminals(matching, chunklen, source_low) || + !helper->CheckTightPhrases(matching, chunklen, source_low)) { return extracts; } int source_back_low = -1, source_back_high = -1; int source_phrase_low = matching[0] - source_sent_start; - int source_phrase_high = matching.back() + chunklen.back() - source_sent_start; + int source_phrase_high = matching.back() + chunklen.back() - + source_sent_start; int target_phrase_low = -1, target_phrase_high = -1; - if (!FindFixPoint(source_phrase_low, source_phrase_high, source_low, - source_high, target_phrase_low, target_phrase_high, - target_low, target_high, source_back_low, source_back_high, - sentence_id, min_gap_size, 0, - max_nonterminals - matching.size() + 1, 1, 1, false)) { + if (!helper->FindFixPoint(source_phrase_low, source_phrase_high, source_low, + source_high, target_phrase_low, target_phrase_high, + target_low, target_high, source_back_low, + source_back_high, sentence_id, min_gap_size, 0, + max_nonterminals - matching.size() + 1, true, true, + false)) { return extracts; } bool met_constraints = true; int num_symbols = phrase.GetNumSymbols(); vector > source_gaps, target_gaps; - if (!CheckGaps(source_gaps, target_gaps, matching, chunklen, source_low, - source_high, target_low, target_high, source_phrase_low, - source_phrase_high, source_back_low, source_back_high, - num_symbols, met_constraints)) { + if (!helper->GetGaps(source_gaps, target_gaps, matching, chunklen, source_low, + source_high, target_low, target_high, source_phrase_low, + source_phrase_high, source_back_low, source_back_high, + num_symbols, met_constraints)) { return extracts; } - bool start_x = source_back_low != source_phrase_low; - bool end_x = source_back_high != source_phrase_high; - Phrase source_phrase = phrase_builder->Extend(phrase, start_x, end_x); + bool starts_with_x = source_back_low != source_phrase_low; + bool ends_with_x = source_back_high != source_phrase_high; + Phrase source_phrase = phrase_builder->Extend( + phrase, starts_with_x, ends_with_x); + unordered_map source_indexes = helper->GetSourceIndexes( + matching, chunklen, starts_with_x); if (met_constraints) { - AddExtracts(extracts, source_phrase, target_gaps, target_low, - target_phrase_low, target_phrase_high, sentence_id); + AddExtracts(extracts, source_phrase, source_indexes, target_gaps, + target_low, target_phrase_low, target_phrase_high, sentence_id); } if (source_gaps.size() >= max_nonterminals || @@ -145,317 +181,24 @@ vector RuleExtractor::ExtractAlignments( for (int i = 0; i < 2; ++i) { for (int j = 1 - i; j < 2; ++j) { - AddNonterminalExtremities(extracts, source_phrase, source_phrase_low, - source_phrase_high, source_back_low, source_back_high, source_low, - source_high, target_low, target_high, target_gaps, sentence_id, i, j); + AddNonterminalExtremities(extracts, matching, chunklen, source_phrase, + source_back_low, source_back_high, source_low, source_high, + target_low, target_high, target_gaps, sentence_id, starts_with_x, + ends_with_x, i, j); } } return extracts; } -void RuleExtractor::GetLinksSpans( - vector& source_low, vector& source_high, - vector& target_low, vector& target_high, int sentence_id) const { - // Ignore end of line markers. - int source_sent_len = source_data_array->GetSentenceStart(sentence_id + 1) - - source_data_array->GetSentenceStart(sentence_id) - 1; - int target_sent_len = target_data_array->GetSentenceStart(sentence_id + 1) - - target_data_array->GetSentenceStart(sentence_id) - 1; - source_low = vector(source_sent_len, -1); - source_high = vector(source_sent_len, -1); - - // TODO(pauldb): Adam Lopez claims this part is really inefficient. See if we - // can speed it up. - target_low = vector(target_sent_len, -1); - target_high = vector(target_sent_len, -1); - const vector >& links = alignment->GetLinks(sentence_id); - for (auto link: links) { - if (source_low[link.first] == -1 || source_low[link.first] > link.second) { - source_low[link.first] = link.second; - } - source_high[link.first] = max(source_high[link.first], link.second + 1); - - if (target_low[link.second] == -1 || target_low[link.second] > link.first) { - target_low[link.second] = link.first; - } - target_high[link.second] = max(target_high[link.second], link.first + 1); - } -} - -bool RuleExtractor::CheckAlignedTerminals(const vector& matching, - const vector& chunklen, - const vector& source_low) const { - if (!require_aligned_terminal) { - return true; - } - - int sentence_id = source_data_array->GetSentenceId(matching[0]); - int source_sent_start = source_data_array->GetSentenceStart(sentence_id); - - int num_aligned_chunks = 0; - for (size_t i = 0; i < chunklen.size(); ++i) { - for (size_t j = 0; j < chunklen[i]; ++j) { - int sent_index = matching[i] - source_sent_start + j; - if (source_low[sent_index] != -1) { - ++num_aligned_chunks; - break; - } - } - } - - if (num_aligned_chunks == 0) { - return false; - } - - return !require_aligned_chunks || num_aligned_chunks == chunklen.size(); -} - -bool RuleExtractor::CheckTightPhrases(const vector& matching, - const vector& chunklen, - const vector& source_low) const { - if (!require_tight_phrases) { - return true; - } - - int sentence_id = source_data_array->GetSentenceId(matching[0]); - int source_sent_start = source_data_array->GetSentenceStart(sentence_id); - for (size_t i = 0; i + 1 < chunklen.size(); ++i) { - int gap_start = matching[i] + chunklen[i] - source_sent_start; - int gap_end = matching[i + 1] - 1 - source_sent_start; - if (source_low[gap_start] == -1 || source_low[gap_end] == -1) { - return false; - } - } - - return true; -} - -bool RuleExtractor::FindFixPoint( - int source_phrase_low, int source_phrase_high, - const vector& source_low, const vector& source_high, - int& target_phrase_low, int& target_phrase_high, - const vector& target_low, const vector& target_high, - int& source_back_low, int& source_back_high, int sentence_id, - int min_source_gap_size, int min_target_gap_size, - int max_new_x, int max_low_x, int max_high_x, - bool allow_arbitrary_expansion) const { - int source_sent_len = source_data_array->GetSentenceStart(sentence_id + 1) - - source_data_array->GetSentenceStart(sentence_id) - 1; - int target_sent_len = target_data_array->GetSentenceStart(sentence_id + 1) - - target_data_array->GetSentenceStart(sentence_id) - 1; - - int prev_target_low = target_phrase_low; - int prev_target_high = target_phrase_high; - FindProjection(source_phrase_low, source_phrase_high, source_low, - source_high, target_phrase_low, target_phrase_high); - - if (target_phrase_low == -1) { - // TODO(pauldb): Low priority corner case inherited from Adam's code: - // If w is unaligned, but we don't require aligned terminals, returning an - // error here prevents the extraction of the allowed rule - // X -> X_1 w X_2 / X_1 X_2 - return false; - } - - if (prev_target_low != -1 && target_phrase_low != prev_target_low) { - if (prev_target_low - target_phrase_low < min_target_gap_size) { - target_phrase_low = prev_target_low - min_target_gap_size; - if (target_phrase_low < 0) { - return false; - } - } - } - - if (prev_target_high != -1 && target_phrase_high != prev_target_high) { - if (target_phrase_high - prev_target_high < min_target_gap_size) { - target_phrase_high = prev_target_high + min_target_gap_size; - if (target_phrase_high > target_sent_len) { - return false; - } - } - } - - if (target_phrase_high - target_phrase_low > max_rule_span) { - return false; - } - - source_back_low = source_back_high = -1; - FindProjection(target_phrase_low, target_phrase_high, target_low, target_high, - source_back_low, source_back_high); - int new_x = 0, new_low_x = 0, new_high_x = 0; - - while (true) { - source_back_low = min(source_back_low, source_phrase_low); - source_back_high = max(source_back_high, source_phrase_high); - - if (source_back_low == source_phrase_low && - source_back_high == source_phrase_high) { - return true; - } - - if (new_low_x >= max_low_x && source_back_low < source_phrase_low) { - // Extension on the left side not allowed. - return false; - } - if (new_high_x >= max_high_x && source_back_high > source_phrase_high) { - // Extension on the right side not allowed. - return false; - } - - // Extend left side. - if (source_back_low < source_phrase_low) { - if (new_x >= max_new_x) { - return false; - } - ++new_x; ++new_low_x; - if (source_phrase_low - source_back_low < min_source_gap_size) { - source_back_low = source_phrase_low - min_source_gap_size; - if (source_back_low < 0) { - return false; - } - } - } - - // Extend right side. - if (source_back_high > source_phrase_high) { - if (new_x >= max_new_x) { - return false; - } - ++new_x; ++new_high_x; - if (source_back_high - source_phrase_high < min_source_gap_size) { - source_back_high = source_phrase_high + min_source_gap_size; - if (source_back_high > source_sent_len) { - return false; - } - } - } - - if (source_back_high - source_back_low > max_rule_span) { - // Rule span too wide. - return false; - } - - prev_target_low = target_phrase_low; - prev_target_high = target_phrase_high; - FindProjection(source_back_low, source_phrase_low, source_low, source_high, - target_phrase_low, target_phrase_high); - FindProjection(source_phrase_high, source_back_high, source_low, - source_high, target_phrase_low, target_phrase_high); - if (prev_target_low == target_phrase_low && - prev_target_high == target_phrase_high) { - return true; - } - - if (!allow_arbitrary_expansion) { - // Arbitrary expansion not allowed. - return false; - } - if (target_phrase_high - target_phrase_low > max_rule_span) { - // Target side too wide. - return false; - } - - source_phrase_low = source_back_low; - source_phrase_high = source_back_high; - FindProjection(target_phrase_low, prev_target_low, target_low, target_high, - source_back_low, source_back_high); - FindProjection(prev_target_high, target_phrase_high, target_low, - target_high, source_back_low, source_back_high); - } - - return false; -} - -void RuleExtractor::FindProjection( - int source_phrase_low, int source_phrase_high, - const vector& source_low, const vector& source_high, - int& target_phrase_low, int& target_phrase_high) const { - for (size_t i = source_phrase_low; i < source_phrase_high; ++i) { - if (source_low[i] != -1) { - if (target_phrase_low == -1 || source_low[i] < target_phrase_low) { - target_phrase_low = source_low[i]; - } - target_phrase_high = max(target_phrase_high, source_high[i]); - } - } -} - -bool RuleExtractor::CheckGaps( - vector >& source_gaps, vector >& target_gaps, - const vector& matching, const vector& chunklen, - const vector& source_low, const vector& source_high, - const vector& target_low, const vector& target_high, - int source_phrase_low, int source_phrase_high, int source_back_low, - int source_back_high, int& num_symbols, bool& met_constraints) const { - int sentence_id = source_data_array->GetSentenceId(matching[0]); - int source_sent_start = source_data_array->GetSentenceStart(sentence_id); - - if (source_back_low < source_phrase_low) { - source_gaps.push_back(make_pair(source_back_low, source_phrase_low)); - if (num_symbols >= max_rule_symbols) { - // Source side contains too many symbols. - return false; - } - ++num_symbols; - if (require_tight_phrases && (source_low[source_back_low] == -1 || - source_low[source_phrase_low - 1] == -1)) { - // Inside edges of preceding gap are not tight. - return false; - } - } else if (require_tight_phrases && source_low[source_phrase_low] == -1) { - // This is not a hard error. We can't extract this phrase, but we might - // still be able to extract a superphrase. - met_constraints = false; - } - - for (size_t i = 0; i + 1 < chunklen.size(); ++i) { - int gap_start = matching[i] + chunklen[i] - source_sent_start; - int gap_end = matching[i + 1] - source_sent_start; - source_gaps.push_back(make_pair(gap_start, gap_end)); - } - - if (source_phrase_high < source_back_high) { - source_gaps.push_back(make_pair(source_phrase_high, source_back_high)); - if (num_symbols >= max_rule_symbols) { - // Source side contains too many symbols. - return false; - } - ++num_symbols; - if (require_tight_phrases && (source_low[source_phrase_high] == -1 || - source_low[source_back_high - 1] == -1)) { - // Inside edges of following gap are not tight. - return false; - } - } else if (require_tight_phrases && - source_low[source_phrase_high - 1] == -1) { - // This is not a hard error. We can't extract this phrase, but we might - // still be able to extract a superphrase. - met_constraints = false; - } - - target_gaps.resize(source_gaps.size(), make_pair(-1, -1)); - for (size_t i = 0; i < source_gaps.size(); ++i) { - if (!FindFixPoint(source_gaps[i].first, source_gaps[i].second, source_low, - source_high, target_gaps[i].first, target_gaps[i].second, - target_low, target_high, source_gaps[i].first, - source_gaps[i].second, sentence_id, 0, 0, 0, 0, 0, - false)) { - // Gap fails integrity check. - return false; - } - } - - return true; -} - void RuleExtractor::AddExtracts( vector& extracts, const Phrase& source_phrase, + const unordered_map& source_indexes, const vector >& target_gaps, const vector& target_low, int target_phrase_low, int target_phrase_high, int sentence_id) const { - vector > target_phrases = ExtractTargetPhrases( + auto target_phrases = target_phrase_extractor->ExtractPhrases( target_gaps, target_low, target_phrase_low, target_phrase_high, - sentence_id); + source_indexes, sentence_id); if (target_phrases.size() > 0) { double pairs_count = 1.0 / target_phrases.size(); @@ -466,147 +209,29 @@ void RuleExtractor::AddExtracts( } } -vector > RuleExtractor::ExtractTargetPhrases( - const vector >& target_gaps, const vector& target_low, - int target_phrase_low, int target_phrase_high, int sentence_id) const { - int target_sent_len = target_data_array->GetSentenceStart(sentence_id + 1) - - target_data_array->GetSentenceStart(sentence_id) - 1; - - vector target_gap_order(target_gaps.size()); - for (size_t i = 0; i < target_gap_order.size(); ++i) { - for (size_t j = 0; j < i; ++j) { - if (target_gaps[target_gap_order[j]] < target_gaps[i]) { - ++target_gap_order[i]; - } else { - ++target_gap_order[j]; - } - } - } - - int target_x_low = target_phrase_low, target_x_high = target_phrase_high; - if (!require_tight_phrases) { - while (target_x_low > 0 && - target_phrase_high - target_x_low < max_rule_span && - target_low[target_x_low - 1] == -1) { - --target_x_low; - } - while (target_x_high + 1 < target_sent_len && - target_x_high - target_phrase_low < max_rule_span && - target_low[target_x_high + 1] == -1) { - ++target_x_high; - } - } - - vector > gaps(target_gaps.size()); - for (size_t i = 0; i < gaps.size(); ++i) { - gaps[i] = target_gaps[target_gap_order[i]]; - if (!require_tight_phrases) { - while (gaps[i].first > target_x_low && - target_low[gaps[i].first] == -1) { - --gaps[i].first; - } - while (gaps[i].second < target_x_high && - target_low[gaps[i].second] == -1) { - ++gaps[i].second; - } - } - } - - vector > ranges(2 * gaps.size() + 2); - ranges.front() = make_pair(target_x_low, target_phrase_low); - ranges.back() = make_pair(target_phrase_high, target_x_high); - for (size_t i = 0; i < gaps.size(); ++i) { - ranges[i * 2 + 1] = make_pair(gaps[i].first, target_gaps[i].first); - ranges[i * 2 + 2] = make_pair(target_gaps[i].second, gaps[i].second); - } - - vector > target_phrases; - vector subpatterns(ranges.size()); - GeneratePhrases(target_phrases, ranges, 0, subpatterns, target_gap_order, - target_phrase_low, target_phrase_high, sentence_id); - return target_phrases; -} - -void RuleExtractor::GeneratePhrases( - vector >& target_phrases, - const vector >& ranges, int index, vector& subpatterns, - const vector& target_gap_order, int target_phrase_low, - int target_phrase_high, int sentence_id) const { - if (index >= ranges.size()) { - if (subpatterns.back() - subpatterns.front() > max_rule_span) { +void RuleExtractor::AddNonterminalExtremities( + vector& extracts, const vector& matching, + const vector& chunklen, const Phrase& source_phrase, + int source_back_low, int source_back_high, const vector& source_low, + const vector& source_high, const vector& target_low, + const vector& target_high, vector > target_gaps, + int sentence_id, int starts_with_x, int ends_with_x, int extend_left, + int extend_right) const { + int source_x_low = source_back_low, source_x_high = source_back_high; + + if (require_tight_phrases) { + if (source_low[source_back_low - extend_left] == -1 || + source_low[source_back_high + extend_right - 1] == -1) { return; } - - vector symbols; - unordered_set target_indexes; - int offset = 1; - if (subpatterns.front() != target_phrase_low) { - offset = 2; - symbols.push_back(vocabulary->GetNonterminalIndex(1)); - } - - int target_sent_start = target_data_array->GetSentenceStart(sentence_id); - for (size_t i = 0; i * 2 < subpatterns.size(); ++i) { - for (size_t j = subpatterns[i * 2]; j < subpatterns[i * 2 + 1]; ++j) { - symbols.push_back(target_data_array->AtIndex(target_sent_start + j)); - target_indexes.insert(j); - } - if (i < target_gap_order.size()) { - symbols.push_back(vocabulary->GetNonterminalIndex( - target_gap_order[i] + offset)); - } - } - - if (subpatterns.back() != target_phrase_high) { - symbols.push_back(target_gap_order.size() + offset); - } - - const vector >& links = alignment->GetLinks(sentence_id); - vector > alignment; - for (pair link: links) { - if (target_indexes.count(link.second)) { - alignment.push_back(link); - } - } - - target_phrases.push_back(make_pair(phrase_builder->Build(symbols), - alignment)); - return; - } - - subpatterns[index] = ranges[index].first; - if (index > 0) { - subpatterns[index] = max(subpatterns[index], subpatterns[index - 1]); } - while (subpatterns[index] <= ranges[index].second) { - subpatterns[index + 1] = max(subpatterns[index], ranges[index + 1].first); - while (subpatterns[index + 1] <= ranges[index + 1].second) { - GeneratePhrases(target_phrases, ranges, index + 2, subpatterns, - target_gap_order, target_phrase_low, target_phrase_high, - sentence_id); - ++subpatterns[index + 1]; - } - ++subpatterns[index]; - } -} -void RuleExtractor::AddNonterminalExtremities( - vector& extracts, const Phrase& source_phrase, - int source_phrase_low, int source_phrase_high, int source_back_low, - int source_back_high, const vector& source_low, - const vector& source_high, const vector& target_low, - const vector& target_high, const vector >& target_gaps, - int sentence_id, int extend_left, int extend_right) const { - int source_x_low = source_phrase_low, source_x_high = source_phrase_high; if (extend_left) { - if (source_back_low != source_phrase_low || - source_phrase_low < min_gap_size || - (require_tight_phrases && (source_low[source_phrase_low - 1] == -1 || - source_low[source_back_high - 1] == -1))) { + if (starts_with_x || source_back_low < min_gap_size) { return; } - source_x_low = source_phrase_low - min_gap_size; + source_x_low = source_back_low - min_gap_size; if (require_tight_phrases) { while (source_x_low >= 0 && source_low[source_x_low] == -1) { --source_x_low; @@ -618,15 +243,11 @@ void RuleExtractor::AddNonterminalExtremities( } if (extend_right) { - int source_sent_len = source_data_array->GetSentenceStart(sentence_id + 1) - - source_data_array->GetSentenceStart(sentence_id) - 1; - if (source_back_high != source_phrase_high || - source_phrase_high + min_gap_size > source_sent_len || - (require_tight_phrases && (source_low[source_phrase_low] == -1 || - source_low[source_phrase_high] == -1))) { + int source_sent_len = source_data_array->GetSentenceLength(sentence_id); + if (ends_with_x || source_back_high + min_gap_size > source_sent_len) { return; } - source_x_high = source_phrase_high + min_gap_size; + source_x_high = source_back_high + min_gap_size; if (require_tight_phrases) { while (source_x_high <= source_sent_len && source_low[source_x_high - 1] == -1) { @@ -639,41 +260,56 @@ void RuleExtractor::AddNonterminalExtremities( } } + int new_nonterminals = extend_left + extend_right; if (source_x_high - source_x_low > max_rule_span || - target_gaps.size() + extend_left + extend_right > max_nonterminals) { + target_gaps.size() + new_nonterminals > max_nonterminals || + source_phrase.GetNumSymbols() + new_nonterminals > max_rule_symbols) { return; } int target_x_low = -1, target_x_high = -1; - if (!FindFixPoint(source_x_low, source_x_high, source_low, source_high, - target_x_low, target_x_high, target_low, target_high, - source_x_low, source_x_high, sentence_id, 1, 1, - extend_left + extend_right, extend_left, extend_right, - true)) { + if (!helper->FindFixPoint(source_x_low, source_x_high, source_low, + source_high, target_x_low, target_x_high, + target_low, target_high, source_x_low, + source_x_high, sentence_id, 1, 1, + new_nonterminals, extend_left, extend_right, + true)) { return; } - int source_gap_low = -1, source_gap_high = -1, target_gap_low = -1, - target_gap_high = -1; - if (extend_left && - ((require_tight_phrases && source_low[source_x_low] == -1) || - !FindFixPoint(source_x_low, source_phrase_low, source_low, source_high, - target_gap_low, target_gap_high, target_low, target_high, - source_gap_low, source_gap_high, sentence_id, - 0, 0, 0, 0, 0, false))) { - return; + if (extend_left) { + int source_gap_low = -1, source_gap_high = -1; + int target_gap_low = -1, target_gap_high = -1; + if ((require_tight_phrases && source_low[source_x_low] == -1) || + !helper->FindFixPoint(source_x_low, source_back_low, source_low, + source_high, target_gap_low, target_gap_high, + target_low, target_high, source_gap_low, + source_gap_high, sentence_id, 0, 0, 0, false, + false, false)) { + return; + } + target_gaps.insert(target_gaps.begin(), + make_pair(target_gap_low, target_gap_high)); } - if (extend_right && - ((require_tight_phrases && source_low[source_x_high - 1] == -1) || - !FindFixPoint(source_phrase_high, source_x_high, source_low, source_high, - target_gap_low, target_gap_high, target_low, target_high, - source_gap_low, source_gap_high, sentence_id, - 0, 0, 0, 0, 0, false))) { - return; + + if (extend_right) { + int target_gap_low = -1, target_gap_high = -1; + int source_gap_low = -1, source_gap_high = -1; + if ((require_tight_phrases && source_low[source_x_high - 1] == -1) || + !helper->FindFixPoint(source_back_high, source_x_high, source_low, + source_high, target_gap_low, target_gap_high, + target_low, target_high, source_gap_low, + source_gap_high, sentence_id, 0, 0, 0, false, + false, false)) { + return; + } + target_gaps.push_back(make_pair(target_gap_low, target_gap_high)); } Phrase new_source_phrase = phrase_builder->Extend(source_phrase, extend_left, extend_right); - AddExtracts(extracts, new_source_phrase, target_gaps, target_low, - target_x_low, target_x_high, sentence_id); + unordered_map source_indexes = helper->GetSourceIndexes( + matching, chunklen, extend_left || starts_with_x); + AddExtracts(extracts, new_source_phrase, source_indexes, target_gaps, + target_low, target_x_low, target_x_high, sentence_id); } diff --git a/extractor/rule_extractor.h b/extractor/rule_extractor.h index f668de24..a087dc6d 100644 --- a/extractor/rule_extractor.h +++ b/extractor/rule_extractor.h @@ -2,6 +2,7 @@ #define _RULE_EXTRACTOR_H_ #include +#include #include #include "phrase.h" @@ -13,8 +14,9 @@ class DataArray; class PhraseBuilder; class PhraseLocation; class Rule; +class RuleExtractorHelper; class Scorer; -class Vocabulary; +class TargetPhraseExtractor; typedef vector > PhraseAlignment; @@ -46,84 +48,56 @@ class RuleExtractor { bool require_aligned_chunks, bool require_tight_phrases); - vector ExtractRules(const Phrase& phrase, - const PhraseLocation& location) const; + // For testing only. + RuleExtractor(shared_ptr source_data_array, + shared_ptr phrase_builder, + shared_ptr scorer, + shared_ptr target_phrase_extractor, + shared_ptr helper, + int max_rule_span, + int min_gap_size, + int max_nonterminals, + int max_rule_symbols, + bool require_tight_phrases); + + virtual ~RuleExtractor(); + + virtual vector ExtractRules(const Phrase& phrase, + const PhraseLocation& location) const; + + protected: + RuleExtractor(); private: vector ExtractAlignments(const Phrase& phrase, const vector& matching) const; - void GetLinksSpans(vector& source_low, vector& source_high, - vector& target_low, vector& target_high, - int sentence_id) const; - - bool CheckAlignedTerminals(const vector& matching, - const vector& chunklen, - const vector& source_low) const; - - bool CheckTightPhrases(const vector& matching, - const vector& chunklen, - const vector& source_low) const; - - bool FindFixPoint( - int source_phrase_start, int source_phrase_end, - const vector& source_low, const vector& source_high, - int& target_phrase_start, int& target_phrase_end, - const vector& target_low, const vector& target_high, - int& source_back_low, int& source_back_high, int sentence_id, - int min_source_gap_size, int min_target_gap_size, - int max_new_x, int max_low_x, int max_high_x, - bool allow_arbitrary_expansion) const; - - void FindProjection( - int source_phrase_start, int source_phrase_end, - const vector& source_low, const vector& source_high, - int& target_phrase_low, int& target_phrase_end) const; - - bool CheckGaps( - vector >& source_gaps, vector >& target_gaps, - const vector& matching, const vector& chunklen, - const vector& source_low, const vector& source_high, - const vector& target_low, const vector& target_high, - int source_phrase_low, int source_phrase_high, int source_back_low, - int source_back_high, int& num_symbols, bool& met_constraints) const; - void AddExtracts( vector& extracts, const Phrase& source_phrase, + const unordered_map& source_indexes, const vector >& target_gaps, const vector& target_low, int target_phrase_low, int target_phrase_high, int sentence_id) const; - vector > ExtractTargetPhrases( - const vector >& target_gaps, const vector& target_low, - int target_phrase_low, int target_phrase_high, int sentence_id) const; - - void GeneratePhrases( - vector >& target_phrases, - const vector >& ranges, int index, - vector& subpatterns, const vector& target_gap_order, - int target_phrase_low, int target_phrase_high, int sentence_id) const; - void AddNonterminalExtremities( - vector& extracts, const Phrase& source_phrase, - int source_phrase_low, int source_phrase_high, int source_back_low, - int source_back_high, const vector& source_low, + vector& extracts, const vector& matching, + const vector& chunklen, const Phrase& source_phrase, + int source_back_low, int source_back_high, const vector& source_low, const vector& source_high, const vector& target_low, - const vector& target_high, - const vector >& target_gaps, int sentence_id, - int extend_left, int extend_right) const; + const vector& target_high, vector > target_gaps, + int sentence_id, int starts_with_x, int ends_with_x, int extend_left, + int extend_right) const; - shared_ptr source_data_array; + private: shared_ptr target_data_array; - shared_ptr alignment; + shared_ptr source_data_array; shared_ptr phrase_builder; shared_ptr scorer; - shared_ptr vocabulary; + shared_ptr target_phrase_extractor; + shared_ptr helper; int max_rule_span; int min_gap_size; int max_nonterminals; int max_rule_symbols; - bool require_aligned_terminal; - bool require_aligned_chunks; bool require_tight_phrases; }; diff --git a/extractor/rule_extractor_helper.cc b/extractor/rule_extractor_helper.cc new file mode 100644 index 00000000..ed6ae3a1 --- /dev/null +++ b/extractor/rule_extractor_helper.cc @@ -0,0 +1,356 @@ +#include "rule_extractor_helper.h" + +#include "data_array.h" +#include "alignment.h" + +RuleExtractorHelper::RuleExtractorHelper( + shared_ptr source_data_array, + shared_ptr target_data_array, + shared_ptr alignment, + int max_rule_span, + int max_rule_symbols, + bool require_aligned_terminal, + bool require_aligned_chunks, + bool require_tight_phrases) : + source_data_array(source_data_array), + target_data_array(target_data_array), + alignment(alignment), + max_rule_span(max_rule_span), + max_rule_symbols(max_rule_symbols), + require_aligned_terminal(require_aligned_terminal), + require_aligned_chunks(require_aligned_chunks), + require_tight_phrases(require_tight_phrases) {} + +RuleExtractorHelper::RuleExtractorHelper() {} + +RuleExtractorHelper::~RuleExtractorHelper() {} + +void RuleExtractorHelper::GetLinksSpans( + vector& source_low, vector& source_high, + vector& target_low, vector& target_high, int sentence_id) const { + int source_sent_len = source_data_array->GetSentenceLength(sentence_id); + int target_sent_len = target_data_array->GetSentenceLength(sentence_id); + source_low = vector(source_sent_len, -1); + source_high = vector(source_sent_len, -1); + + // TODO(pauldb): Adam Lopez claims this part is really inefficient. See if we + // can speed it up. + target_low = vector(target_sent_len, -1); + target_high = vector(target_sent_len, -1); + vector > links = alignment->GetLinks(sentence_id); + for (auto link: links) { + if (source_low[link.first] == -1 || source_low[link.first] > link.second) { + source_low[link.first] = link.second; + } + source_high[link.first] = max(source_high[link.first], link.second + 1); + + if (target_low[link.second] == -1 || target_low[link.second] > link.first) { + target_low[link.second] = link.first; + } + target_high[link.second] = max(target_high[link.second], link.first + 1); + } +} + +bool RuleExtractorHelper::CheckAlignedTerminals( + const vector& matching, + const vector& chunklen, + const vector& source_low) const { + if (!require_aligned_terminal) { + return true; + } + + int sentence_id = source_data_array->GetSentenceId(matching[0]); + int source_sent_start = source_data_array->GetSentenceStart(sentence_id); + + int num_aligned_chunks = 0; + for (size_t i = 0; i < chunklen.size(); ++i) { + for (size_t j = 0; j < chunklen[i]; ++j) { + int sent_index = matching[i] - source_sent_start + j; + if (source_low[sent_index] != -1) { + ++num_aligned_chunks; + break; + } + } + } + + if (num_aligned_chunks == 0) { + return false; + } + + return !require_aligned_chunks || num_aligned_chunks == chunklen.size(); +} + +bool RuleExtractorHelper::CheckTightPhrases( + const vector& matching, + const vector& chunklen, + const vector& source_low) const { + if (!require_tight_phrases) { + return true; + } + + int sentence_id = source_data_array->GetSentenceId(matching[0]); + int source_sent_start = source_data_array->GetSentenceStart(sentence_id); + for (size_t i = 0; i + 1 < chunklen.size(); ++i) { + int gap_start = matching[i] + chunklen[i] - source_sent_start; + int gap_end = matching[i + 1] - 1 - source_sent_start; + if (source_low[gap_start] == -1 || source_low[gap_end] == -1) { + return false; + } + } + + return true; +} + +bool RuleExtractorHelper::FindFixPoint( + int source_phrase_low, int source_phrase_high, + const vector& source_low, const vector& source_high, + int& target_phrase_low, int& target_phrase_high, + const vector& target_low, const vector& target_high, + int& source_back_low, int& source_back_high, int sentence_id, + int min_source_gap_size, int min_target_gap_size, + int max_new_x, bool allow_low_x, bool allow_high_x, + bool allow_arbitrary_expansion) const { + int prev_target_low = target_phrase_low; + int prev_target_high = target_phrase_high; + + FindProjection(source_phrase_low, source_phrase_high, source_low, + source_high, target_phrase_low, target_phrase_high); + + if (target_phrase_low == -1) { + // TODO(pauldb): Low priority corner case inherited from Adam's code: + // If w is unaligned, but we don't require aligned terminals, returning an + // error here prevents the extraction of the allowed rule + // X -> X_1 w X_2 / X_1 X_2 + return false; + } + + int source_sent_len = source_data_array->GetSentenceLength(sentence_id); + int target_sent_len = target_data_array->GetSentenceLength(sentence_id); + if (prev_target_low != -1 && target_phrase_low != prev_target_low) { + if (prev_target_low - target_phrase_low < min_target_gap_size) { + target_phrase_low = prev_target_low - min_target_gap_size; + if (target_phrase_low < 0) { + return false; + } + } + } + + if (prev_target_high != -1 && target_phrase_high != prev_target_high) { + if (target_phrase_high - prev_target_high < min_target_gap_size) { + target_phrase_high = prev_target_high + min_target_gap_size; + if (target_phrase_high > target_sent_len) { + return false; + } + } + } + + if (target_phrase_high - target_phrase_low > max_rule_span) { + return false; + } + + source_back_low = source_back_high = -1; + FindProjection(target_phrase_low, target_phrase_high, target_low, target_high, + source_back_low, source_back_high); + int new_x = 0; + bool new_low_x = false, new_high_x = false; + while (true) { + source_back_low = min(source_back_low, source_phrase_low); + source_back_high = max(source_back_high, source_phrase_high); + + if (source_back_low == source_phrase_low && + source_back_high == source_phrase_high) { + return true; + } + + if (!allow_low_x && source_back_low < source_phrase_low) { + // Extension on the left side not allowed. + return false; + } + if (!allow_high_x && source_back_high > source_phrase_high) { + // Extension on the right side not allowed. + return false; + } + + // Extend left side. + if (source_back_low < source_phrase_low) { + if (new_low_x == false) { + if (new_x >= max_new_x) { + return false; + } + new_low_x = true; + ++new_x; + } + if (source_phrase_low - source_back_low < min_source_gap_size) { + source_back_low = source_phrase_low - min_source_gap_size; + if (source_back_low < 0) { + return false; + } + } + } + + // Extend right side. + if (source_back_high > source_phrase_high) { + if (new_high_x == false) { + if (new_x >= max_new_x) { + return false; + } + new_high_x = true; + ++new_x; + } + if (source_back_high - source_phrase_high < min_source_gap_size) { + source_back_high = source_phrase_high + min_source_gap_size; + if (source_back_high > source_sent_len) { + return false; + } + } + } + + if (source_back_high - source_back_low > max_rule_span) { + // Rule span too wide. + return false; + } + + prev_target_low = target_phrase_low; + prev_target_high = target_phrase_high; + FindProjection(source_back_low, source_phrase_low, source_low, source_high, + target_phrase_low, target_phrase_high); + FindProjection(source_phrase_high, source_back_high, source_low, + source_high, target_phrase_low, target_phrase_high); + if (prev_target_low == target_phrase_low && + prev_target_high == target_phrase_high) { + return true; + } + + if (!allow_arbitrary_expansion) { + // Arbitrary expansion not allowed. + return false; + } + if (target_phrase_high - target_phrase_low > max_rule_span) { + // Target side too wide. + return false; + } + + source_phrase_low = source_back_low; + source_phrase_high = source_back_high; + FindProjection(target_phrase_low, prev_target_low, target_low, target_high, + source_back_low, source_back_high); + FindProjection(prev_target_high, target_phrase_high, target_low, + target_high, source_back_low, source_back_high); + } + + return false; +} + +void RuleExtractorHelper::FindProjection( + int source_phrase_low, int source_phrase_high, + const vector& source_low, const vector& source_high, + int& target_phrase_low, int& target_phrase_high) const { + for (size_t i = source_phrase_low; i < source_phrase_high; ++i) { + if (source_low[i] != -1) { + if (target_phrase_low == -1 || source_low[i] < target_phrase_low) { + target_phrase_low = source_low[i]; + } + target_phrase_high = max(target_phrase_high, source_high[i]); + } + } +} + +bool RuleExtractorHelper::GetGaps( + vector >& source_gaps, vector >& target_gaps, + const vector& matching, const vector& chunklen, + const vector& source_low, const vector& source_high, + const vector& target_low, const vector& target_high, + int source_phrase_low, int source_phrase_high, int source_back_low, + int source_back_high, int& num_symbols, bool& met_constraints) const { + int sentence_id = source_data_array->GetSentenceId(matching[0]); + int source_sent_start = source_data_array->GetSentenceStart(sentence_id); + + if (source_back_low < source_phrase_low) { + source_gaps.push_back(make_pair(source_back_low, source_phrase_low)); + if (num_symbols >= max_rule_symbols) { + // Source side contains too many symbols. + return false; + } + ++num_symbols; + if (require_tight_phrases && (source_low[source_back_low] == -1 || + source_low[source_phrase_low - 1] == -1)) { + // Inside edges of preceding gap are not tight. + return false; + } + } else if (require_tight_phrases && source_low[source_phrase_low] == -1) { + // This is not a hard error. We can't extract this phrase, but we might + // still be able to extract a superphrase. + met_constraints = false; + } + + for (size_t i = 0; i + 1 < chunklen.size(); ++i) { + int gap_start = matching[i] + chunklen[i] - source_sent_start; + int gap_end = matching[i + 1] - source_sent_start; + source_gaps.push_back(make_pair(gap_start, gap_end)); + } + + if (source_phrase_high < source_back_high) { + source_gaps.push_back(make_pair(source_phrase_high, source_back_high)); + if (num_symbols >= max_rule_symbols) { + // Source side contains too many symbols. + return false; + } + ++num_symbols; + if (require_tight_phrases && (source_low[source_phrase_high] == -1 || + source_low[source_back_high - 1] == -1)) { + // Inside edges of following gap are not tight. + return false; + } + } else if (require_tight_phrases && + source_low[source_phrase_high - 1] == -1) { + // This is not a hard error. We can't extract this phrase, but we might + // still be able to extract a superphrase. + met_constraints = false; + } + + target_gaps.resize(source_gaps.size(), make_pair(-1, -1)); + for (size_t i = 0; i < source_gaps.size(); ++i) { + if (!FindFixPoint(source_gaps[i].first, source_gaps[i].second, source_low, + source_high, target_gaps[i].first, target_gaps[i].second, + target_low, target_high, source_gaps[i].first, + source_gaps[i].second, sentence_id, 0, 0, 0, false, false, + false)) { + // Gap fails integrity check. + return false; + } + } + + return true; +} + +vector RuleExtractorHelper::GetGapOrder( + const vector >& gaps) const { + vector gap_order(gaps.size()); + for (size_t i = 0; i < gap_order.size(); ++i) { + for (size_t j = 0; j < i; ++j) { + if (gaps[gap_order[j]] < gaps[i]) { + ++gap_order[i]; + } else { + ++gap_order[j]; + } + } + } + return gap_order; +} + +unordered_map RuleExtractorHelper::GetSourceIndexes( + const vector& matching, const vector& chunklen, + int starts_with_x) const { + unordered_map source_indexes; + int sentence_id = source_data_array->GetSentenceId(matching[0]); + int source_sent_start = source_data_array->GetSentenceStart(sentence_id); + int num_symbols = starts_with_x; + for (size_t i = 0; i < matching.size(); ++i) { + for (size_t j = 0; j < chunklen[i]; ++j) { + source_indexes[matching[i] + j - source_sent_start] = num_symbols; + ++num_symbols; + } + ++num_symbols; + } + return source_indexes; +} diff --git a/extractor/rule_extractor_helper.h b/extractor/rule_extractor_helper.h new file mode 100644 index 00000000..3478bfc8 --- /dev/null +++ b/extractor/rule_extractor_helper.h @@ -0,0 +1,82 @@ +#ifndef _RULE_EXTRACTOR_HELPER_H_ +#define _RULE_EXTRACTOR_HELPER_H_ + +#include +#include +#include + +using namespace std; + +class Alignment; +class DataArray; + +class RuleExtractorHelper { + public: + RuleExtractorHelper(shared_ptr source_data_array, + shared_ptr target_data_array, + shared_ptr alignment, + int max_rule_span, + int max_rule_symbols, + bool require_aligned_terminal, + bool require_aligned_chunks, + bool require_tight_phrases); + + virtual ~RuleExtractorHelper(); + + virtual void GetLinksSpans(vector& source_low, vector& source_high, + vector& target_low, vector& target_high, + int sentence_id) const; + + virtual bool CheckAlignedTerminals(const vector& matching, + const vector& chunklen, + const vector& source_low) const; + + virtual bool CheckTightPhrases(const vector& matching, + const vector& chunklen, + const vector& source_low) const; + + virtual bool FindFixPoint( + int source_phrase_low, int source_phrase_high, + const vector& source_low, const vector& source_high, + int& target_phrase_low, int& target_phrase_high, + const vector& target_low, const vector& target_high, + int& source_back_low, int& source_back_high, int sentence_id, + int min_source_gap_size, int min_target_gap_size, + int max_new_x, bool allow_low_x, bool allow_high_x, + bool allow_arbitrary_expansion) const; + + virtual bool GetGaps( + vector >& source_gaps, vector >& target_gaps, + const vector& matching, const vector& chunklen, + const vector& source_low, const vector& source_high, + const vector& target_low, const vector& target_high, + int source_phrase_low, int source_phrase_high, int source_back_low, + int source_back_high, int& num_symbols, bool& met_constraints) const; + + virtual vector GetGapOrder(const vector >& gaps) const; + + // TODO(pauldb): Add unit tests. + virtual unordered_map GetSourceIndexes( + const vector& matching, const vector& chunklen, + int starts_with_x) const; + + protected: + RuleExtractorHelper(); + + private: + void FindProjection( + int source_phrase_low, int source_phrase_high, + const vector& source_low, const vector& source_high, + int& target_phrase_low, int& target_phrase_high) const; + + shared_ptr source_data_array; + shared_ptr target_data_array; + shared_ptr alignment; + int max_rule_span; + int max_rule_symbols; + bool require_aligned_terminal; + bool require_aligned_chunks; + bool require_tight_phrases; +}; + +#endif diff --git a/extractor/rule_extractor_helper_test.cc b/extractor/rule_extractor_helper_test.cc new file mode 100644 index 00000000..29213312 --- /dev/null +++ b/extractor/rule_extractor_helper_test.cc @@ -0,0 +1,622 @@ +#include + +#include + +#include "mocks/mock_alignment.h" +#include "mocks/mock_data_array.h" +#include "rule_extractor_helper.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class RuleExtractorHelperTest : public Test { + protected: + virtual void SetUp() { + source_data_array = make_shared(); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(12)); + EXPECT_CALL(*source_data_array, GetSentenceId(_)) + .WillRepeatedly(Return(5)); + EXPECT_CALL(*source_data_array, GetSentenceStart(_)) + .WillRepeatedly(Return(10)); + + target_data_array = make_shared(); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(12)); + + vector > links = { + make_pair(0, 0), make_pair(0, 1), make_pair(2, 2), make_pair(3, 1) + }; + alignment = make_shared(); + EXPECT_CALL(*alignment, GetLinks(_)).WillRepeatedly(Return(links)); + } + + shared_ptr source_data_array; + shared_ptr target_data_array; + shared_ptr alignment; + shared_ptr helper; +}; + +TEST_F(RuleExtractorHelperTest, TestGetLinksSpans) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(4)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(3)); + + vector source_low, source_high, target_low, target_high; + helper->GetLinksSpans(source_low, source_high, target_low, target_high, 0); + + vector expected_source_low = {0, -1, 2, 1}; + EXPECT_EQ(expected_source_low, source_low); + vector expected_source_high = {2, -1, 3, 2}; + EXPECT_EQ(expected_source_high, source_high); + vector expected_target_low = {0, 0, 2}; + EXPECT_EQ(expected_target_low, target_low); + vector expected_target_high = {1, 4, 3}; + EXPECT_EQ(expected_target_high, target_high); +} + +TEST_F(RuleExtractorHelperTest, TestCheckAlignedFalse) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, false, false, true); + EXPECT_CALL(*source_data_array, GetSentenceId(_)).Times(0); + EXPECT_CALL(*source_data_array, GetSentenceStart(_)).Times(0); + + vector matching, chunklen, source_low; + EXPECT_TRUE(helper->CheckAlignedTerminals(matching, chunklen, source_low)); +} + +TEST_F(RuleExtractorHelperTest, TestCheckAlignedTerminal) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, false, true); + + vector matching = {10, 12}; + vector chunklen = {1, 3}; + vector source_low = {-1, 1, -1, 3, -1}; + EXPECT_TRUE(helper->CheckAlignedTerminals(matching, chunklen, source_low)); + source_low = {-1, 1, -1, -1, -1}; + EXPECT_FALSE(helper->CheckAlignedTerminals(matching, chunklen, source_low)); +} + +TEST_F(RuleExtractorHelperTest, TestCheckAlignedChunks) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + + vector matching = {10, 12}; + vector chunklen = {1, 3}; + vector source_low = {2, 1, -1, 3, -1}; + EXPECT_TRUE(helper->CheckAlignedTerminals(matching, chunklen, source_low)); + source_low = {-1, 1, -1, 3, -1}; + EXPECT_FALSE(helper->CheckAlignedTerminals(matching, chunklen, source_low)); + source_low = {2, 1, -1, -1, -1}; + EXPECT_FALSE(helper->CheckAlignedTerminals(matching, chunklen, source_low)); +} + + +TEST_F(RuleExtractorHelperTest, TestCheckTightPhrasesFalse) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, false); + EXPECT_CALL(*source_data_array, GetSentenceId(_)).Times(0); + EXPECT_CALL(*source_data_array, GetSentenceStart(_)).Times(0); + + vector matching, chunklen, source_low; + EXPECT_TRUE(helper->CheckTightPhrases(matching, chunklen, source_low)); +} + +TEST_F(RuleExtractorHelperTest, TestCheckTightPhrases) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + + vector matching = {10, 14, 18}; + vector chunklen = {2, 3, 1}; + // No missing links. + vector source_low = {0, 1, 2, 3, 4, 5, 6, 7, 8}; + EXPECT_TRUE(helper->CheckTightPhrases(matching, chunklen, source_low)); + + // Missing link at the beginning or ending of a gap. + source_low = {0, 1, -1, 3, 4, 5, 6, 7, 8}; + EXPECT_FALSE(helper->CheckTightPhrases(matching, chunklen, source_low)); + source_low = {0, 1, 2, -1, 4, 5, 6, 7, 8}; + EXPECT_FALSE(helper->CheckTightPhrases(matching, chunklen, source_low)); + source_low = {0, 1, 2, 3, 4, 5, 6, -1, 8}; + EXPECT_FALSE(helper->CheckTightPhrases(matching, chunklen, source_low)); + + // Missing link inside the gap. + chunklen = {1, 3, 1}; + source_low = {0, 1, -1, 3, 4, 5, 6, 7, 8}; + EXPECT_TRUE(helper->CheckTightPhrases(matching, chunklen, source_low)); +} + +TEST_F(RuleExtractorHelperTest, TestFindFixPointBadEdgeCase) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + + vector source_low = {0, -1, 2}; + vector source_high = {1, -1, 3}; + vector target_low = {0, -1, 2}; + vector target_high = {1, -1, 3}; + int source_phrase_low = 1, source_phrase_high = 2; + int source_back_low, source_back_high; + int target_phrase_low = -1, target_phrase_high = 1; + + // This should be in fact true. See comment about the inherited bug. + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 0, 0, + 0, false, false, false)); +} + +TEST_F(RuleExtractorHelperTest, TestFindFixPointTargetSentenceOutOfBounds) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(3)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(3)); + + vector source_low = {0, 0, 2}; + vector source_high = {1, 2, 3}; + vector target_low = {0, 1, 2}; + vector target_high = {2, 2, 3}; + int source_phrase_low = 1, source_phrase_high = 2; + int source_back_low, source_back_high; + int target_phrase_low = 1, target_phrase_high = 2; + + // Extend out of sentence to left. + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 2, 2, + 0, false, false, false)); + source_low = {0, 1, 2}; + source_high = {1, 3, 3}; + target_low = {0, 1, 1}; + target_high = {1, 2, 3}; + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 2, 2, + 0, false, false, false)); +} + +TEST_F(RuleExtractorHelperTest, TestFindFixPointTargetTooWide) { + helper = make_shared(source_data_array, + target_data_array, alignment, 5, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + + vector source_low = {0, 0, 0, 0, 0, 0, 0}; + vector source_high = {7, 7, 7, 7, 7, 7, 7}; + vector target_low = {0, -1, -1, -1, -1, -1, 0}; + vector target_high = {7, -1, -1, -1, -1, -1, 7}; + int source_phrase_low = 2, source_phrase_high = 5; + int source_back_low, source_back_high; + int target_phrase_low = -1, target_phrase_high = -1; + + // Projection is too wide. + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 1, 1, + 0, false, false, false)); +} + +TEST_F(RuleExtractorHelperTest, TestFindFixPoint) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + + vector source_low = {1, 1, 1, 3, 4, 5, 5}; + vector source_high = {2, 2, 3, 4, 6, 6, 6}; + vector target_low = {-1, 0, 2, 3, 4, 4, -1}; + vector target_high = {-1, 3, 3, 4, 5, 7, -1}; + int source_phrase_low = 2, source_phrase_high = 5; + int source_back_low, source_back_high; + int target_phrase_low = 2, target_phrase_high = 5; + + EXPECT_TRUE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 1, 1, 1, + 2, true, true, false)); + EXPECT_EQ(1, target_phrase_low); + EXPECT_EQ(6, target_phrase_high); + EXPECT_EQ(0, source_back_low); + EXPECT_EQ(7, source_back_high); + + source_low = {0, -1, 1, 3, 4, -1, 6}; + source_high = {1, -1, 3, 4, 6, -1, 7}; + target_low = {0, 2, 2, 3, 4, 4, 6}; + target_high = {1, 3, 3, 4, 5, 5, 7}; + source_phrase_low = 2, source_phrase_high = 5; + target_phrase_low = -1, target_phrase_high = -1; + EXPECT_TRUE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 1, 1, 1, + 2, true, true, false)); + EXPECT_EQ(1, target_phrase_low); + EXPECT_EQ(6, target_phrase_high); + EXPECT_EQ(2, source_back_low); + EXPECT_EQ(5, source_back_high); +} + +TEST_F(RuleExtractorHelperTest, TestFindFixPointExtensionsNotAllowed) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(3)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(3)); + + vector source_low = {0, 0, 2}; + vector source_high = {1, 2, 3}; + vector target_low = {0, 1, 2}; + vector target_high = {2, 2, 3}; + int source_phrase_low = 1, source_phrase_high = 2; + int source_back_low, source_back_high; + int target_phrase_low = -1, target_phrase_high = -1; + + // Extension on the left side not allowed. + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 1, 1, + 1, false, true, false)); + // Extension on the left side is allowed, but we can't add anymore X. + target_phrase_low = -1, target_phrase_high = -1; + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 1, 1, + 0, true, true, false)); + source_low = {0, 1, 2}; + source_high = {1, 3, 3}; + target_low = {0, 1, 1}; + target_high = {1, 2, 3}; + // Extension on the right side not allowed. + target_phrase_low = -1, target_phrase_high = -1; + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 1, 1, + 1, true, false, false)); + // Extension on the right side is allowed, but we can't add anymore X. + target_phrase_low = -1, target_phrase_high = -1; + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 1, 1, + 0, true, true, false)); +} + +TEST_F(RuleExtractorHelperTest, TestFindFixPointSourceSentenceOutOfBounds) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(3)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(3)); + + vector source_low = {0, 0, 2}; + vector source_high = {1, 2, 3}; + vector target_low = {0, 1, 2}; + vector target_high = {2, 2, 3}; + int source_phrase_low = 1, source_phrase_high = 2; + int source_back_low, source_back_high; + int target_phrase_low = 1, target_phrase_high = 2; + // Extend out of sentence to left. + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 2, 1, + 1, true, true, false)); + source_low = {0, 1, 2}; + source_high = {1, 3, 3}; + target_low = {0, 1, 1}; + target_high = {1, 2, 3}; + target_phrase_low = 1, target_phrase_high = 2; + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 2, 1, + 1, true, true, false)); +} + +TEST_F(RuleExtractorHelperTest, TestFindFixPointTargetSourceWide) { + helper = make_shared(source_data_array, + target_data_array, alignment, 5, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + + vector source_low = {2, -1, 2, 3, 4, -1, 4}; + vector source_high = {3, -1, 3, 4, 5, -1, 5}; + vector target_low = {-1, -1, 0, 3, 4, -1, -1}; + vector target_high = {-1, -1, 3, 4, 7, -1, -1}; + int source_phrase_low = 2, source_phrase_high = 5; + int source_back_low, source_back_high; + int target_phrase_low = -1, target_phrase_high = -1; + + // Second projection (on source side) is too wide. + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 1, 1, + 2, true, true, false)); +} + +TEST_F(RuleExtractorHelperTest, TestFindFixPointArbitraryExpansion) { + helper = make_shared(source_data_array, + target_data_array, alignment, 20, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(11)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(11)); + + vector source_low = {1, 1, 2, 3, 4, 5, 6, 7, 7, 8, 9}; + vector source_high = {2, 3, 4, 5, 5, 6, 7, 8, 9, 10, 10}; + vector target_low = {-1, 0, 1, 2, 3, 5, 6, 7, 8, 9, -1}; + vector target_high = {-1, 2, 3, 4, 5, 6, 8, 9, 10, 11, -1}; + int source_phrase_low = 4, source_phrase_high = 7; + int source_back_low, source_back_high; + int target_phrase_low = -1, target_phrase_high = -1; + EXPECT_FALSE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 1, 1, + 10, true, true, false)); + + source_phrase_low = 4, source_phrase_high = 7; + target_phrase_low = -1, target_phrase_high = -1; + EXPECT_TRUE(helper->FindFixPoint(source_phrase_low, source_phrase_high, + source_low, source_high, target_phrase_low, + target_phrase_high, target_low, target_high, + source_back_low, source_back_high, 0, 1, 1, + 10, true, true, true)); +} + +TEST_F(RuleExtractorHelperTest, TestGetGapOrder) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + + vector > gaps = + {make_pair(0, 3), make_pair(5, 8), make_pair(11, 12), make_pair(15, 17)}; + vector expected_gap_order = {0, 1, 2, 3}; + EXPECT_EQ(expected_gap_order, helper->GetGapOrder(gaps)); + + gaps = {make_pair(15, 17), make_pair(8, 9), make_pair(5, 6), make_pair(0, 3)}; + expected_gap_order = {3, 2, 1, 0}; + EXPECT_EQ(expected_gap_order, helper->GetGapOrder(gaps)); + + gaps = {make_pair(8, 9), make_pair(5, 6), make_pair(0, 3), make_pair(15, 17)}; + expected_gap_order = {2, 1, 0, 3}; + EXPECT_EQ(expected_gap_order, helper->GetGapOrder(gaps)); +} + +TEST_F(RuleExtractorHelperTest, TestGetGapsExceedNumSymbols) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + + bool met_constraints = true; + vector source_low = {1, 1, 2, 3, 4, 5, 6}; + vector source_high = {2, 2, 3, 4, 5, 6, 7}; + vector target_low = {-1, 0, 2, 3, 4, 5, 6}; + vector target_high = {-1, 2, 3, 4, 5, 6, 7}; + int source_phrase_low = 1, source_phrase_high = 6; + int source_back_low = 0, source_back_high = 6; + vector matching = {11, 13, 15}; + vector chunklen = {1, 1, 1}; + vector > source_gaps, target_gaps; + int num_symbols = 5; + EXPECT_FALSE(helper->GetGaps(source_gaps, target_gaps, matching, chunklen, + source_low, source_high, target_low, target_high, + source_phrase_low, source_phrase_high, + source_back_low, source_back_high, num_symbols, + met_constraints)); + + source_low = {0, 1, 2, 3, 4, 5, 5}; + source_high = {1, 2, 3, 4, 5, 6, 6}; + target_low = {0, 1, 2, 3, 4, 5, -1}; + target_high = {1, 2, 3, 4, 5, 7, -1}; + source_phrase_low = 1, source_phrase_high = 6; + source_back_low = 1, source_back_high = 7; + num_symbols = 5; + EXPECT_FALSE(helper->GetGaps(source_gaps, target_gaps, matching, chunklen, + source_low, source_high, target_low, target_high, + source_phrase_low, source_phrase_high, + source_back_low, source_back_high, num_symbols, + met_constraints)); +} + +TEST_F(RuleExtractorHelperTest, TestGetGapsExtensionsNotTight) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 7, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + + bool met_constraints = true; + vector source_low = {-1, 1, 2, 3, 4, 5, -1}; + vector source_high = {-1, 2, 3, 4, 5, 6, -1}; + vector target_low = {-1, 1, 2, 3, 4, 5, -1}; + vector target_high = {-1, 2, 3, 4, 5, 6, -1}; + int source_phrase_low = 1, source_phrase_high = 6; + int source_back_low = 0, source_back_high = 6; + vector matching = {11, 13, 15}; + vector chunklen = {1, 1, 1}; + vector > source_gaps, target_gaps; + int num_symbols = 5; + EXPECT_FALSE(helper->GetGaps(source_gaps, target_gaps, matching, chunklen, + source_low, source_high, target_low, target_high, + source_phrase_low, source_phrase_high, + source_back_low, source_back_high, num_symbols, + met_constraints)); + + source_phrase_low = 1, source_phrase_high = 6; + source_back_low = 1, source_back_high = 7; + num_symbols = 5; + EXPECT_FALSE(helper->GetGaps(source_gaps, target_gaps, matching, chunklen, + source_low, source_high, target_low, target_high, + source_phrase_low, source_phrase_high, + source_back_low, source_back_high, num_symbols, + met_constraints)); +} + +TEST_F(RuleExtractorHelperTest, TestGetGapsNotTightExtremities) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 7, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + + bool met_constraints = true; + vector source_low = {-1, -1, 2, 3, 4, 5, 6}; + vector source_high = {-1, -1, 3, 4, 5, 6, 7}; + vector target_low = {-1, -1, 2, 3, 4, 5, 6}; + vector target_high = {-1, -1, 3, 4, 5, 6, 7}; + int source_phrase_low = 1, source_phrase_high = 6; + int source_back_low = 1, source_back_high = 6; + vector matching = {11, 13, 15}; + vector chunklen = {1, 1, 1}; + vector > source_gaps, target_gaps; + int num_symbols = 5; + EXPECT_TRUE(helper->GetGaps(source_gaps, target_gaps, matching, chunklen, + source_low, source_high, target_low, target_high, + source_phrase_low, source_phrase_high, + source_back_low, source_back_high, num_symbols, + met_constraints)); + EXPECT_FALSE(met_constraints); + vector > expected_gaps = {make_pair(2, 3), make_pair(4, 5)}; + EXPECT_EQ(expected_gaps, source_gaps); + EXPECT_EQ(expected_gaps, target_gaps); + + source_low = {-1, 1, 2, 3, 4, -1, 6}; + source_high = {-1, 2, 3, 4, 5, -1, 7}; + target_low = {-1, 1, 2, 3, 4, -1, 6}; + target_high = {-1, 2, 3, 4, 5, -1, 7}; + met_constraints = true; + source_gaps.clear(); + target_gaps.clear(); + EXPECT_TRUE(helper->GetGaps(source_gaps, target_gaps, matching, chunklen, + source_low, source_high, target_low, target_high, + source_phrase_low, source_phrase_high, + source_back_low, source_back_high, num_symbols, + met_constraints)); + EXPECT_FALSE(met_constraints); + EXPECT_EQ(expected_gaps, source_gaps); + EXPECT_EQ(expected_gaps, target_gaps); +} + +TEST_F(RuleExtractorHelperTest, TestGetGapsWithExtensions) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + + bool met_constraints = true; + vector source_low = {-1, 5, 2, 3, 4, 1, -1}; + vector source_high = {-1, 6, 3, 4, 5, 2, -1}; + vector target_low = {-1, 5, 2, 3, 4, 1, -1}; + vector target_high = {-1, 6, 3, 4, 5, 2, -1}; + int source_phrase_low = 2, source_phrase_high = 5; + int source_back_low = 1, source_back_high = 6; + vector matching = {12, 14}; + vector chunklen = {1, 1}; + vector > source_gaps, target_gaps; + int num_symbols = 3; + EXPECT_TRUE(helper->GetGaps(source_gaps, target_gaps, matching, chunklen, + source_low, source_high, target_low, target_high, + source_phrase_low, source_phrase_high, + source_back_low, source_back_high, num_symbols, + met_constraints)); + vector > expected_source_gaps = { + make_pair(1, 2), make_pair(3, 4), make_pair(5, 6) + }; + EXPECT_EQ(expected_source_gaps, source_gaps); + vector > expected_target_gaps = { + make_pair(5, 6), make_pair(3, 4), make_pair(1, 2) + }; + EXPECT_EQ(expected_target_gaps, target_gaps); +} + +TEST_F(RuleExtractorHelperTest, TestGetGaps) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + + bool met_constraints = true; + vector source_low = {-1, 1, 4, 3, 2, 5, -1}; + vector source_high = {-1, 2, 5, 4, 3, 6, -1}; + vector target_low = {-1, 1, 4, 3, 2, 5, -1}; + vector target_high = {-1, 2, 5, 4, 3, 6, -1}; + int source_phrase_low = 1, source_phrase_high = 6; + int source_back_low = 1, source_back_high = 6; + vector matching = {11, 13, 15}; + vector chunklen = {1, 1, 1}; + vector > source_gaps, target_gaps; + int num_symbols = 5; + EXPECT_TRUE(helper->GetGaps(source_gaps, target_gaps, matching, chunklen, + source_low, source_high, target_low, target_high, + source_phrase_low, source_phrase_high, + source_back_low, source_back_high, num_symbols, + met_constraints)); + vector > expected_source_gaps = { + make_pair(2, 3), make_pair(4, 5) + }; + EXPECT_EQ(expected_source_gaps, source_gaps); + vector > expected_target_gaps = { + make_pair(4, 5), make_pair(2, 3) + }; + EXPECT_EQ(expected_target_gaps, target_gaps); +} + +TEST_F(RuleExtractorHelperTest, TestGetGapIntegrityChecksFailed) { + helper = make_shared(source_data_array, + target_data_array, alignment, 10, 5, true, true, true); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + EXPECT_CALL(*target_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(7)); + + bool met_constraints = true; + vector source_low = {-1, 3, 2, 3, 4, 3, -1}; + vector source_high = {-1, 4, 3, 4, 5, 4, -1}; + vector target_low = {-1, -1, 2, 1, 4, -1, -1}; + vector target_high = {-1, -1, 3, 6, 5, -1, -1}; + int source_phrase_low = 2, source_phrase_high = 5; + int source_back_low = 2, source_back_high = 5; + vector matching = {12, 14}; + vector chunklen = {1, 1}; + vector > source_gaps, target_gaps; + int num_symbols = 3; + EXPECT_FALSE(helper->GetGaps(source_gaps, target_gaps, matching, chunklen, + source_low, source_high, target_low, target_high, + source_phrase_low, source_phrase_high, + source_back_low, source_back_high, num_symbols, + met_constraints)); +} + +} // namespace diff --git a/extractor/rule_extractor_test.cc b/extractor/rule_extractor_test.cc new file mode 100644 index 00000000..0be44d4d --- /dev/null +++ b/extractor/rule_extractor_test.cc @@ -0,0 +1,166 @@ +#include + +#include + +#include "mocks/mock_alignment.h" +#include "mocks/mock_data_array.h" +#include "mocks/mock_rule_extractor_helper.h" +#include "mocks/mock_scorer.h" +#include "mocks/mock_target_phrase_extractor.h" +#include "mocks/mock_vocabulary.h" +#include "phrase.h" +#include "phrase_builder.h" +#include "phrase_location.h" +#include "rule_extractor.h" +#include "rule.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class RuleExtractorTest : public Test { + protected: + virtual void SetUp() { + source_data_array = make_shared(); + EXPECT_CALL(*source_data_array, GetSentenceId(_)) + .WillRepeatedly(Return(0)); + EXPECT_CALL(*source_data_array, GetSentenceStart(_)) + .WillRepeatedly(Return(0)); + EXPECT_CALL(*source_data_array, GetSentenceLength(_)) + .WillRepeatedly(Return(10)); + + helper = make_shared(); + EXPECT_CALL(*helper, CheckAlignedTerminals(_, _, _)) + .WillRepeatedly(Return(true)); + EXPECT_CALL(*helper, CheckTightPhrases(_, _, _)) + .WillRepeatedly(Return(true)); + unordered_map source_indexes; + EXPECT_CALL(*helper, GetSourceIndexes(_, _, _)) + .WillRepeatedly(Return(source_indexes)); + + vocabulary = make_shared(); + EXPECT_CALL(*vocabulary, GetTerminalValue(87)) + .WillRepeatedly(Return("a")); + phrase_builder = make_shared(vocabulary); + vector symbols = {87}; + Phrase target_phrase = phrase_builder->Build(symbols); + PhraseAlignment phrase_alignment = {make_pair(0, 0)}; + + target_phrase_extractor = make_shared(); + vector > target_phrases = { + make_pair(target_phrase, phrase_alignment) + }; + EXPECT_CALL(*target_phrase_extractor, ExtractPhrases(_, _, _, _, _, _)) + .WillRepeatedly(Return(target_phrases)); + + scorer = make_shared(); + vector scores = {0.3, 7.2}; + EXPECT_CALL(*scorer, Score(_)).WillRepeatedly(Return(scores)); + + extractor = make_shared(source_data_array, phrase_builder, + scorer, target_phrase_extractor, helper, 10, 1, 3, 5, false); + } + + shared_ptr source_data_array; + shared_ptr vocabulary; + shared_ptr phrase_builder; + shared_ptr helper; + shared_ptr scorer; + shared_ptr target_phrase_extractor; + shared_ptr extractor; +}; + +TEST_F(RuleExtractorTest, TestExtractRulesAlignedTerminalsFail) { + vector symbols = {87}; + Phrase phrase = phrase_builder->Build(symbols); + vector matching = {2}; + PhraseLocation phrase_location(matching, 1); + EXPECT_CALL(*helper, GetLinksSpans(_, _, _, _, _)).Times(1); + EXPECT_CALL(*helper, CheckAlignedTerminals(_, _, _)) + .WillRepeatedly(Return(false)); + vector rules = extractor->ExtractRules(phrase, phrase_location); + EXPECT_EQ(0, rules.size()); +} + +TEST_F(RuleExtractorTest, TestExtractRulesTightPhrasesFail) { + vector symbols = {87}; + Phrase phrase = phrase_builder->Build(symbols); + vector matching = {2}; + PhraseLocation phrase_location(matching, 1); + EXPECT_CALL(*helper, GetLinksSpans(_, _, _, _, _)).Times(1); + EXPECT_CALL(*helper, CheckTightPhrases(_, _, _)) + .WillRepeatedly(Return(false)); + vector rules = extractor->ExtractRules(phrase, phrase_location); + EXPECT_EQ(0, rules.size()); +} + +TEST_F(RuleExtractorTest, TestExtractRulesNoFixPoint) { + vector symbols = {87}; + Phrase phrase = phrase_builder->Build(symbols); + vector matching = {2}; + PhraseLocation phrase_location(matching, 1); + + EXPECT_CALL(*helper, GetLinksSpans(_, _, _, _, _)).Times(1); + // Set FindFixPoint to return false. + vector > gaps; + helper->SetUp(0, 0, 0, 0, false, gaps, gaps, 0, true, true); + + vector rules = extractor->ExtractRules(phrase, phrase_location); + EXPECT_EQ(0, rules.size()); +} + +TEST_F(RuleExtractorTest, TestExtractRulesGapsFail) { + vector symbols = {87}; + Phrase phrase = phrase_builder->Build(symbols); + vector matching = {2}; + PhraseLocation phrase_location(matching, 1); + + EXPECT_CALL(*helper, GetLinksSpans(_, _, _, _, _)).Times(1); + // Set CheckGaps to return false. + vector > gaps; + helper->SetUp(0, 0, 0, 0, true, gaps, gaps, 0, true, false); + + vector rules = extractor->ExtractRules(phrase, phrase_location); + EXPECT_EQ(0, rules.size()); +} + +TEST_F(RuleExtractorTest, TestExtractRulesNoExtremities) { + vector symbols = {87}; + Phrase phrase = phrase_builder->Build(symbols); + vector matching = {2}; + PhraseLocation phrase_location(matching, 1); + + EXPECT_CALL(*helper, GetLinksSpans(_, _, _, _, _)).Times(1); + vector > gaps(3); + // Set FindFixPoint to return true. The number of gaps equals the number of + // nonterminals, so we won't add any extremities. + helper->SetUp(0, 0, 0, 0, true, gaps, gaps, 0, true, true); + + vector rules = extractor->ExtractRules(phrase, phrase_location); + EXPECT_EQ(1, rules.size()); +} + +TEST_F(RuleExtractorTest, TestExtractRulesAddExtremities) { + vector symbols = {87}; + Phrase phrase = phrase_builder->Build(symbols); + vector matching = {2}; + PhraseLocation phrase_location(matching, 1); + + vector links(10, -1); + EXPECT_CALL(*helper, GetLinksSpans(_, _, _, _, _)).WillOnce(DoAll( + SetArgReferee<0>(links), + SetArgReferee<1>(links), + SetArgReferee<2>(links), + SetArgReferee<3>(links))); + + vector > gaps; + // Set FindFixPoint to return true. The number of gaps equals the number of + // nonterminals, so we won't add any extremities. + helper->SetUp(0, 0, 2, 3, true, gaps, gaps, 0, true, true); + + vector rules = extractor->ExtractRules(phrase, phrase_location); + EXPECT_EQ(4, rules.size()); +} + +} // namespace diff --git a/extractor/rule_factory.cc b/extractor/rule_factory.cc index c22f9b48..374a0db1 100644 --- a/extractor/rule_factory.cc +++ b/extractor/rule_factory.cc @@ -1,6 +1,6 @@ #include "rule_factory.h" -#include +#include #include #include #include @@ -18,7 +18,9 @@ #include "vocabulary.h" using namespace std; -using namespace tr1; +using namespace std::chrono; + +typedef high_resolution_clock Clock; struct State { State(int start, int end, const vector& phrase, @@ -68,8 +70,44 @@ HieroCachingRuleFactory::HieroCachingRuleFactory( sampler = make_shared(source_suffix_array, max_samples); } +HieroCachingRuleFactory::HieroCachingRuleFactory( + shared_ptr finder, + shared_ptr intersector, + shared_ptr phrase_builder, + shared_ptr rule_extractor, + shared_ptr vocabulary, + shared_ptr sampler, + shared_ptr scorer, + int min_gap_size, + int max_rule_span, + int max_nonterminals, + int max_chunks, + int max_rule_symbols) : + matchings_finder(finder), + intersector(intersector), + phrase_builder(phrase_builder), + rule_extractor(rule_extractor), + vocabulary(vocabulary), + sampler(sampler), + scorer(scorer), + min_gap_size(min_gap_size), + max_rule_span(max_rule_span), + max_nonterminals(max_nonterminals), + max_chunks(max_chunks), + max_rule_symbols(max_rule_symbols) {} + +HieroCachingRuleFactory::HieroCachingRuleFactory() {} + +HieroCachingRuleFactory::~HieroCachingRuleFactory() {} Grammar HieroCachingRuleFactory::GetGrammar(const vector& word_ids) { + intersector->binary_merge_time = 0; + intersector->linear_merge_time = 0; + intersector->sort_time = 0; + Clock::time_point start_time = Clock::now(); + double total_extract_time = 0; + double total_intersect_time = 0; + double total_lookup_time = 0; // Clear cache for every new sentence. trie.Reset(); shared_ptr root = trie.GetRoot(); @@ -107,34 +145,42 @@ Grammar HieroCachingRuleFactory::GetGrammar(const vector& word_ids) { } if (RequiresLookup(node, word_id)) { - shared_ptr next_suffix_link = - node->suffix_link->GetChild(word_id); + shared_ptr next_suffix_link = node->suffix_link == NULL ? + trie.GetRoot() : node->suffix_link->GetChild(word_id); if (state.starts_with_x) { // If the phrase starts with a non terminal, we simply use the matchings // from the suffix link. - next_node = shared_ptr(new TrieNode( - next_suffix_link, next_phrase, next_suffix_link->matchings)); + next_node = make_shared( + next_suffix_link, next_phrase, next_suffix_link->matchings); } else { PhraseLocation phrase_location; if (next_phrase.Arity() > 0) { + Clock::time_point intersect_start_time = Clock::now(); phrase_location = intersector->Intersect( node->phrase, node->matchings, next_suffix_link->phrase, next_suffix_link->matchings, next_phrase); + Clock::time_point intersect_stop_time = Clock::now(); + total_intersect_time += duration_cast( + intersect_stop_time - intersect_start_time).count(); } else { + Clock::time_point lookup_start_time = Clock::now(); phrase_location = matchings_finder->Find( node->matchings, vocabulary->GetTerminalValue(word_id), state.phrase.size()); + Clock::time_point lookup_stop_time = Clock::now(); + total_lookup_time += duration_cast( + lookup_stop_time - lookup_start_time).count(); } if (phrase_location.IsEmpty()) { continue; } - next_node = shared_ptr(new TrieNode( - next_suffix_link, next_phrase, phrase_location)); + next_node = make_shared( + next_suffix_link, next_phrase, phrase_location); } node->AddChild(word_id, next_node); @@ -143,12 +189,16 @@ Grammar HieroCachingRuleFactory::GetGrammar(const vector& word_ids) { AddTrailingNonterminal(phrase, next_phrase, next_node, state.starts_with_x); + Clock::time_point extract_start_time = Clock::now(); if (!state.starts_with_x) { PhraseLocation sample = sampler->Sample(next_node->matchings); vector new_rules = rule_extractor->ExtractRules(next_phrase, sample); rules.insert(rules.end(), new_rules.begin(), new_rules.end()); } + Clock::time_point extract_stop_time = Clock::now(); + total_extract_time += duration_cast( + extract_stop_time - extract_start_time).count(); } else { next_node = node->GetChild(word_id); } @@ -160,6 +210,16 @@ Grammar HieroCachingRuleFactory::GetGrammar(const vector& word_ids) { } } + Clock::time_point stop_time = Clock::now(); + milliseconds ms = duration_cast(stop_time - start_time); + cerr << "Total time for rule lookup, extraction, and scoring = " + << ms.count() / 1000.0 << endl; + cerr << "Extract time = " << total_extract_time / 1000.0 << endl; + cerr << "Intersect time = " << total_intersect_time / 1000.0 << endl; + cerr << "Sort time = " << intersector->sort_time / 1000.0 << endl; + cerr << "Linear merge time = " << intersector->linear_merge_time / 1000.0 << endl; + cerr << "Binary merge time = " << intersector->binary_merge_time / 1000.0 << endl; + // cerr << "Lookup time = " << total_lookup_time / 1000.0 << endl; return Grammar(rules, scorer->GetFeatureNames()); } @@ -192,12 +252,12 @@ void HieroCachingRuleFactory::AddTrailingNonterminal( Phrase var_phrase = phrase_builder->Build(symbols); int suffix_var_id = vocabulary->GetNonterminalIndex( - prefix.Arity() + starts_with_x == 0); + prefix.Arity() + (starts_with_x == 0)); shared_ptr var_suffix_link = prefix_node->suffix_link->GetChild(suffix_var_id); - prefix_node->AddChild(var_id, shared_ptr(new TrieNode( - var_suffix_link, var_phrase, prefix_node->matchings))); + prefix_node->AddChild(var_id, make_shared( + var_suffix_link, var_phrase, prefix_node->matchings)); } vector HieroCachingRuleFactory::ExtendState( @@ -216,7 +276,7 @@ vector HieroCachingRuleFactory::ExtendState( new_states.push_back(State(state.start, state.end + 1, symbols, state.subpatterns_start, node, state.starts_with_x)); - int num_subpatterns = phrase.Arity() + state.starts_with_x == 0; + int num_subpatterns = phrase.Arity() + (state.starts_with_x == 0); if (symbols.size() + 1 >= max_rule_symbols || phrase.Arity() >= max_nonterminals || num_subpatterns >= max_chunks) { diff --git a/extractor/rule_factory.h b/extractor/rule_factory.h index a47b6d16..cf344667 100644 --- a/extractor/rule_factory.h +++ b/extractor/rule_factory.h @@ -40,7 +40,27 @@ class HieroCachingRuleFactory { bool use_beaza_yates, bool require_tight_phrases); - Grammar GetGrammar(const vector& word_ids); + // For testing only. + HieroCachingRuleFactory( + shared_ptr finder, + shared_ptr intersector, + shared_ptr phrase_builder, + shared_ptr rule_extractor, + shared_ptr vocabulary, + shared_ptr sampler, + shared_ptr scorer, + int min_gap_size, + int max_rule_span, + int max_nonterminals, + int max_chunks, + int max_rule_symbols); + + virtual ~HieroCachingRuleFactory(); + + virtual Grammar GetGrammar(const vector& word_ids); + + protected: + HieroCachingRuleFactory(); private: bool CannotHaveMatchings(shared_ptr node, int word_id); diff --git a/extractor/rule_factory_test.cc b/extractor/rule_factory_test.cc new file mode 100644 index 00000000..d6fbab74 --- /dev/null +++ b/extractor/rule_factory_test.cc @@ -0,0 +1,98 @@ +#include + +#include +#include +#include + +#include "grammar.h" +#include "mocks/mock_intersector.h" +#include "mocks/mock_matchings_finder.h" +#include "mocks/mock_rule_extractor.h" +#include "mocks/mock_sampler.h" +#include "mocks/mock_scorer.h" +#include "mocks/mock_vocabulary.h" +#include "phrase_builder.h" +#include "phrase_location.h" +#include "rule_factory.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class RuleFactoryTest : public Test { + protected: + virtual void SetUp() { + finder = make_shared(); + intersector = make_shared(); + + vocabulary = make_shared(); + EXPECT_CALL(*vocabulary, GetTerminalValue(2)).WillRepeatedly(Return("a")); + EXPECT_CALL(*vocabulary, GetTerminalValue(3)).WillRepeatedly(Return("b")); + EXPECT_CALL(*vocabulary, GetTerminalValue(4)).WillRepeatedly(Return("c")); + + phrase_builder = make_shared(vocabulary); + + scorer = make_shared(); + feature_names = {"f1"}; + EXPECT_CALL(*scorer, GetFeatureNames()) + .WillRepeatedly(Return(feature_names)); + + sampler = make_shared(); + EXPECT_CALL(*sampler, Sample(_)) + .WillRepeatedly(Return(PhraseLocation(0, 1))); + + Phrase phrase; + vector scores = {0.5}; + vector > phrase_alignment = {make_pair(0, 0)}; + vector rules = {Rule(phrase, phrase, scores, phrase_alignment)}; + extractor = make_shared(); + EXPECT_CALL(*extractor, ExtractRules(_, _)) + .WillRepeatedly(Return(rules)); + + factory = make_shared(finder, intersector, + phrase_builder, extractor, vocabulary, sampler, scorer, 1, 10, 2, 3, 5); + } + + vector feature_names; + shared_ptr finder; + shared_ptr intersector; + shared_ptr vocabulary; + shared_ptr phrase_builder; + shared_ptr scorer; + shared_ptr sampler; + shared_ptr extractor; + shared_ptr factory; +}; + +TEST_F(RuleFactoryTest, TestGetGrammarDifferentWords) { + EXPECT_CALL(*finder, Find(_, _, _)) + .Times(6) + .WillRepeatedly(Return(PhraseLocation(0, 1))); + + EXPECT_CALL(*intersector, Intersect(_, _, _, _, _)) + .Times(1) + .WillRepeatedly(Return(PhraseLocation(0, 1))); + + vector word_ids = {2, 3, 4}; + Grammar grammar = factory->GetGrammar(word_ids); + EXPECT_EQ(feature_names, grammar.GetFeatureNames()); + EXPECT_EQ(7, grammar.GetRules().size()); +} + +TEST_F(RuleFactoryTest, TestGetGrammarRepeatingWords) { + EXPECT_CALL(*finder, Find(_, _, _)) + .Times(12) + .WillRepeatedly(Return(PhraseLocation(0, 1))); + + EXPECT_CALL(*intersector, Intersect(_, _, _, _, _)) + .Times(16) + .WillRepeatedly(Return(PhraseLocation(0, 1))); + + vector word_ids = {2, 3, 4, 2, 3}; + Grammar grammar = factory->GetGrammar(word_ids); + EXPECT_EQ(feature_names, grammar.GetFeatureNames()); + EXPECT_EQ(28, grammar.GetRules().size()); +} + +} // namespace diff --git a/extractor/run_extractor.cc b/extractor/run_extractor.cc index 37a9cba0..ed30e6fe 100644 --- a/extractor/run_extractor.cc +++ b/extractor/run_extractor.cc @@ -114,8 +114,8 @@ int main(int argc, char** argv) { make_shared(), make_shared(), make_shared(), - make_shared(table), make_shared(table), + make_shared(table), make_shared(), make_shared() }; @@ -138,6 +138,10 @@ int main(int argc, char** argv) { int grammar_id = 0; fs::path grammar_path = vm["grammars"].as(); + if (!fs::is_directory(grammar_path)) { + fs::create_directory(grammar_path); + } + string sentence, delimiter = "|||"; while (getline(cin, sentence)) { string suffix = ""; @@ -148,7 +152,8 @@ int main(int argc, char** argv) { } Grammar grammar = extractor.GetGrammar(sentence); - fs::path grammar_file = grammar_path / to_string(grammar_id); + string file_name = "grammar." + to_string(grammar_id); + fs::path grammar_file = grammar_path / file_name; ofstream output(grammar_file.c_str()); output << grammar; diff --git a/extractor/sample_alignment.txt b/extractor/sample_alignment.txt new file mode 100644 index 00000000..80b446a4 --- /dev/null +++ b/extractor/sample_alignment.txt @@ -0,0 +1,2 @@ +0-0 1-1 2-2 +1-0 2-1 diff --git a/extractor/sampler.cc b/extractor/sampler.cc index d8e0f49e..5067ca8a 100644 --- a/extractor/sampler.cc +++ b/extractor/sampler.cc @@ -6,6 +6,10 @@ Sampler::Sampler(shared_ptr suffix_array, int max_samples) : suffix_array(suffix_array), max_samples(max_samples) {} +Sampler::Sampler() {} + +Sampler::~Sampler() {} + PhraseLocation Sampler::Sample(const PhraseLocation& location) const { vector sample; int num_subpatterns; @@ -32,5 +36,6 @@ PhraseLocation Sampler::Sample(const PhraseLocation& location) const { } int Sampler::Round(double x) const { - return x + 0.5; + // TODO(pauldb): Remove EPS. + return x + 0.5 + 1e-8; } diff --git a/extractor/sampler.h b/extractor/sampler.h index 3b3e3a4d..9cf321fb 100644 --- a/extractor/sampler.h +++ b/extractor/sampler.h @@ -12,7 +12,12 @@ class Sampler { public: Sampler(shared_ptr suffix_array, int max_samples); - PhraseLocation Sample(const PhraseLocation& location) const; + virtual ~Sampler(); + + virtual PhraseLocation Sample(const PhraseLocation& location) const; + + protected: + Sampler(); private: int Round(double x) const; diff --git a/extractor/scorer.cc b/extractor/scorer.cc index c87e179d..f28b3181 100644 --- a/extractor/scorer.cc +++ b/extractor/scorer.cc @@ -5,6 +5,10 @@ Scorer::Scorer(const vector >& features) : features(features) {} +Scorer::Scorer() {} + +Scorer::~Scorer() {} + vector Scorer::Score(const FeatureContext& context) const { vector scores; for (auto feature: features) { diff --git a/extractor/scorer.h b/extractor/scorer.h index 5b328fb4..ba71a6ee 100644 --- a/extractor/scorer.h +++ b/extractor/scorer.h @@ -14,9 +14,14 @@ class Scorer { public: Scorer(const vector >& features); - vector Score(const FeatureContext& context) const; + virtual ~Scorer(); - vector GetFeatureNames() const; + virtual vector Score(const FeatureContext& context) const; + + virtual vector GetFeatureNames() const; + + protected: + Scorer(); private: vector > features; diff --git a/extractor/scorer_test.cc b/extractor/scorer_test.cc new file mode 100644 index 00000000..56a85762 --- /dev/null +++ b/extractor/scorer_test.cc @@ -0,0 +1,47 @@ +#include + +#include +#include +#include + +#include "mocks/mock_feature.h" +#include "scorer.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class ScorerTest : public Test { + protected: + virtual void SetUp() { + feature1 = make_shared(); + EXPECT_CALL(*feature1, Score(_)).WillRepeatedly(Return(0.5)); + EXPECT_CALL(*feature1, GetName()).WillRepeatedly(Return("f1")); + + feature2 = make_shared(); + EXPECT_CALL(*feature2, Score(_)).WillRepeatedly(Return(-1.3)); + EXPECT_CALL(*feature2, GetName()).WillRepeatedly(Return("f2")); + + vector > features = {feature1, feature2}; + scorer = make_shared(features); + } + + shared_ptr feature1; + shared_ptr feature2; + shared_ptr scorer; +}; + +TEST_F(ScorerTest, TestScore) { + vector expected_scores = {0.5, -1.3}; + Phrase phrase; + FeatureContext context(phrase, phrase, 0.3, 2, 11); + EXPECT_EQ(expected_scores, scorer->Score(context)); +} + +TEST_F(ScorerTest, TestGetNames) { + vector expected_names = {"f1", "f2"}; + EXPECT_EQ(expected_names, scorer->GetFeatureNames()); +} + +} // namespace diff --git a/extractor/suffix_array.cc b/extractor/suffix_array.cc index d13eacd5..9815996f 100644 --- a/extractor/suffix_array.cc +++ b/extractor/suffix_array.cc @@ -22,9 +22,9 @@ SuffixArray::~SuffixArray() {} void SuffixArray::BuildSuffixArray() { vector groups = data_array->GetData(); groups.reserve(groups.size() + 1); - groups.push_back(data_array->GetVocabularySize()); + groups.push_back(DataArray::NULL_WORD); suffix_array.resize(groups.size()); - word_start.resize(data_array->GetVocabularySize() + 2); + word_start.resize(data_array->GetVocabularySize() + 1); InitialBucketSort(groups); @@ -112,6 +112,8 @@ void SuffixArray::TernaryQuicksort(int left, int right, int step, } } + TernaryQuicksort(left, mid_left - 1, step, groups); + if (mid_left == mid_right) { groups[suffix_array[mid_left]] = mid_left; suffix_array[mid_left] = -1; @@ -121,7 +123,6 @@ void SuffixArray::TernaryQuicksort(int left, int right, int step, } } - TernaryQuicksort(left, mid_left - 1, step, groups); TernaryQuicksort(mid_right + 1, right, step, groups); } @@ -201,7 +202,7 @@ int SuffixArray::LookupRangeStart(int low, int high, int word_id, int result = high; while (low < high) { int middle = low + (high - low) / 2; - if (suffix_array[middle] + offset < data_array->GetSize() && + if (suffix_array[middle] + offset >= data_array->GetSize() || data_array->AtIndex(suffix_array[middle] + offset) < word_id) { low = middle + 1; } else { diff --git a/extractor/suffix_array_test.cc b/extractor/suffix_array_test.cc index d891933c..60295567 100644 --- a/extractor/suffix_array_test.cc +++ b/extractor/suffix_array_test.cc @@ -14,10 +14,10 @@ namespace { class SuffixArrayTest : public Test { protected: virtual void SetUp() { - data = vector{5, 3, 0, 1, 3, 4, 2, 3, 5, 5, 3, 0, 1}; + data = {6, 4, 1, 2, 4, 5, 3, 4, 6, 6, 4, 1, 2}; data_array = make_shared(); EXPECT_CALL(*data_array, GetData()).WillRepeatedly(ReturnRef(data)); - EXPECT_CALL(*data_array, GetVocabularySize()).WillRepeatedly(Return(6)); + EXPECT_CALL(*data_array, GetVocabularySize()).WillRepeatedly(Return(7)); EXPECT_CALL(*data_array, GetSize()).WillRepeatedly(Return(13)); suffix_array = make_shared(data_array); } @@ -33,14 +33,15 @@ TEST_F(SuffixArrayTest, TestData) { } TEST_F(SuffixArrayTest, TestBuildSuffixArray) { - vector expected_suffix_array{2, 11, 3, 12, 6, 1, 10, 4, 7, 5, 0, 9, 8}; + vector expected_suffix_array = + {13, 11, 2, 12, 3, 6, 10, 1, 4, 7, 5, 9, 0, 8}; for (size_t i = 0; i < expected_suffix_array.size(); ++i) { EXPECT_EQ(expected_suffix_array[i], suffix_array->GetSuffix(i)); } } TEST_F(SuffixArrayTest, TestBuildLCP) { - vector expected_lcp{-1, 2, 0, 1, 0, 0, 3, 1, 1, 0, 0, 4, 1, 0}; + vector expected_lcp = {-1, 0, 2, 0, 1, 0, 0, 3, 1, 1, 0, 0, 4, 1}; EXPECT_EQ(expected_lcp, suffix_array->BuildLCPArray()); } @@ -50,26 +51,26 @@ TEST_F(SuffixArrayTest, TestLookup) { } EXPECT_CALL(*data_array, HasWord("word1")).WillRepeatedly(Return(true)); - EXPECT_CALL(*data_array, GetWordId("word1")).WillRepeatedly(Return(5)); - EXPECT_EQ(PhraseLocation(10, 13), suffix_array->Lookup(0, 14, "word1", 0)); + EXPECT_CALL(*data_array, GetWordId("word1")).WillRepeatedly(Return(6)); + EXPECT_EQ(PhraseLocation(11, 14), suffix_array->Lookup(0, 14, "word1", 0)); EXPECT_CALL(*data_array, HasWord("word2")).WillRepeatedly(Return(false)); EXPECT_EQ(PhraseLocation(0, 0), suffix_array->Lookup(0, 14, "word2", 0)); EXPECT_CALL(*data_array, HasWord("word3")).WillRepeatedly(Return(true)); - EXPECT_CALL(*data_array, GetWordId("word3")).WillRepeatedly(Return(3)); - EXPECT_EQ(PhraseLocation(10, 12), suffix_array->Lookup(10, 13, "word3", 1)); + EXPECT_CALL(*data_array, GetWordId("word3")).WillRepeatedly(Return(4)); + EXPECT_EQ(PhraseLocation(11, 13), suffix_array->Lookup(11, 14, "word3", 1)); EXPECT_CALL(*data_array, HasWord("word4")).WillRepeatedly(Return(true)); - EXPECT_CALL(*data_array, GetWordId("word4")).WillRepeatedly(Return(0)); - EXPECT_EQ(PhraseLocation(10, 12), suffix_array->Lookup(10, 12, "word4", 2)); + EXPECT_CALL(*data_array, GetWordId("word4")).WillRepeatedly(Return(1)); + EXPECT_EQ(PhraseLocation(11, 13), suffix_array->Lookup(11, 13, "word4", 2)); EXPECT_CALL(*data_array, HasWord("word5")).WillRepeatedly(Return(true)); - EXPECT_CALL(*data_array, GetWordId("word5")).WillRepeatedly(Return(1)); - EXPECT_EQ(PhraseLocation(10, 12), suffix_array->Lookup(10, 12, "word5", 3)); + EXPECT_CALL(*data_array, GetWordId("word5")).WillRepeatedly(Return(2)); + EXPECT_EQ(PhraseLocation(11, 13), suffix_array->Lookup(11, 13, "word5", 3)); - EXPECT_EQ(PhraseLocation(10, 11), suffix_array->Lookup(10, 12, "word3", 4)); - EXPECT_EQ(PhraseLocation(10, 10), suffix_array->Lookup(10, 12, "word5", 1)); + EXPECT_EQ(PhraseLocation(12, 13), suffix_array->Lookup(11, 13, "word3", 4)); + EXPECT_EQ(PhraseLocation(11, 11), suffix_array->Lookup(11, 13, "word5", 1)); } } // namespace diff --git a/extractor/target_phrase_extractor.cc b/extractor/target_phrase_extractor.cc new file mode 100644 index 00000000..ac583953 --- /dev/null +++ b/extractor/target_phrase_extractor.cc @@ -0,0 +1,144 @@ +#include "target_phrase_extractor.h" + +#include + +#include "alignment.h" +#include "data_array.h" +#include "phrase.h" +#include "phrase_builder.h" +#include "rule_extractor_helper.h" +#include "vocabulary.h" + +using namespace std; + +TargetPhraseExtractor::TargetPhraseExtractor( + shared_ptr target_data_array, + shared_ptr alignment, + shared_ptr phrase_builder, + shared_ptr helper, + shared_ptr vocabulary, + int max_rule_span, + bool require_tight_phrases) : + target_data_array(target_data_array), + alignment(alignment), + phrase_builder(phrase_builder), + helper(helper), + vocabulary(vocabulary), + max_rule_span(max_rule_span), + require_tight_phrases(require_tight_phrases) {} + +TargetPhraseExtractor::TargetPhraseExtractor() {} + +TargetPhraseExtractor::~TargetPhraseExtractor() {} + +vector > TargetPhraseExtractor::ExtractPhrases( + const vector >& target_gaps, const vector& target_low, + int target_phrase_low, int target_phrase_high, + const unordered_map& source_indexes, int sentence_id) const { + int target_sent_len = target_data_array->GetSentenceLength(sentence_id); + + vector target_gap_order = helper->GetGapOrder(target_gaps); + + int target_x_low = target_phrase_low, target_x_high = target_phrase_high; + if (!require_tight_phrases) { + while (target_x_low > 0 && + target_phrase_high - target_x_low < max_rule_span && + target_low[target_x_low - 1] == -1) { + --target_x_low; + } + while (target_x_high < target_sent_len && + target_x_high - target_phrase_low < max_rule_span && + target_low[target_x_high] == -1) { + ++target_x_high; + } + } + + vector > gaps(target_gaps.size()); + for (size_t i = 0; i < gaps.size(); ++i) { + gaps[i] = target_gaps[target_gap_order[i]]; + if (!require_tight_phrases) { + while (gaps[i].first > target_x_low && + target_low[gaps[i].first - 1] == -1) { + --gaps[i].first; + } + while (gaps[i].second < target_x_high && + target_low[gaps[i].second] == -1) { + ++gaps[i].second; + } + } + } + + vector > ranges(2 * gaps.size() + 2); + ranges.front() = make_pair(target_x_low, target_phrase_low); + ranges.back() = make_pair(target_phrase_high, target_x_high); + for (size_t i = 0; i < gaps.size(); ++i) { + int j = target_gap_order[i]; + ranges[i * 2 + 1] = make_pair(gaps[i].first, target_gaps[j].first); + ranges[i * 2 + 2] = make_pair(target_gaps[j].second, gaps[i].second); + } + + vector > target_phrases; + vector subpatterns(ranges.size()); + GeneratePhrases(target_phrases, ranges, 0, subpatterns, target_gap_order, + target_phrase_low, target_phrase_high, source_indexes, + sentence_id); + return target_phrases; +} + +void TargetPhraseExtractor::GeneratePhrases( + vector >& target_phrases, + const vector >& ranges, int index, vector& subpatterns, + const vector& target_gap_order, int target_phrase_low, + int target_phrase_high, const unordered_map& source_indexes, + int sentence_id) const { + if (index >= ranges.size()) { + if (subpatterns.back() - subpatterns.front() > max_rule_span) { + return; + } + + vector symbols; + unordered_map target_indexes; + + int target_sent_start = target_data_array->GetSentenceStart(sentence_id); + for (size_t i = 0; i * 2 < subpatterns.size(); ++i) { + for (size_t j = subpatterns[i * 2]; j < subpatterns[i * 2 + 1]; ++j) { + target_indexes[j] = symbols.size(); + string target_word = target_data_array->GetWordAtIndex( + target_sent_start + j); + symbols.push_back(vocabulary->GetTerminalIndex(target_word)); + } + if (i < target_gap_order.size()) { + symbols.push_back(vocabulary->GetNonterminalIndex( + target_gap_order[i] + 1)); + } + } + + vector > links = alignment->GetLinks(sentence_id); + vector > alignment; + for (pair link: links) { + if (target_indexes.count(link.second)) { + alignment.push_back(make_pair(source_indexes.find(link.first)->second, + target_indexes[link.second])); + } + } + + Phrase target_phrase = phrase_builder->Build(symbols); + target_phrases.push_back(make_pair(target_phrase, alignment)); + return; + } + + subpatterns[index] = ranges[index].first; + if (index > 0) { + subpatterns[index] = max(subpatterns[index], subpatterns[index - 1]); + } + while (subpatterns[index] <= ranges[index].second) { + subpatterns[index + 1] = max(subpatterns[index], ranges[index + 1].first); + while (subpatterns[index + 1] <= ranges[index + 1].second) { + GeneratePhrases(target_phrases, ranges, index + 2, subpatterns, + target_gap_order, target_phrase_low, target_phrase_high, + source_indexes, sentence_id); + ++subpatterns[index + 1]; + } + ++subpatterns[index]; + } +} diff --git a/extractor/target_phrase_extractor.h b/extractor/target_phrase_extractor.h new file mode 100644 index 00000000..134f24cc --- /dev/null +++ b/extractor/target_phrase_extractor.h @@ -0,0 +1,56 @@ +#ifndef _TARGET_PHRASE_EXTRACTOR_H_ +#define _TARGET_PHRASE_EXTRACTOR_H_ + +#include +#include +#include + +using namespace std; + +class Alignment; +class DataArray; +class Phrase; +class PhraseBuilder; +class RuleExtractorHelper; +class Vocabulary; + +typedef vector > PhraseAlignment; + +class TargetPhraseExtractor { + public: + TargetPhraseExtractor(shared_ptr target_data_array, + shared_ptr alignment, + shared_ptr phrase_builder, + shared_ptr helper, + shared_ptr vocabulary, + int max_rule_span, + bool require_tight_phrases); + + virtual ~TargetPhraseExtractor(); + + virtual vector > ExtractPhrases( + const vector >& target_gaps, const vector& target_low, + int target_phrase_low, int target_phrase_high, + const unordered_map& source_indexes, int sentence_id) const; + + protected: + TargetPhraseExtractor(); + + private: + void GeneratePhrases( + vector >& target_phrases, + const vector >& ranges, int index, + vector& subpatterns, const vector& target_gap_order, + int target_phrase_low, int target_phrase_high, + const unordered_map& source_indexes, int sentence_id) const; + + shared_ptr target_data_array; + shared_ptr alignment; + shared_ptr phrase_builder; + shared_ptr helper; + shared_ptr vocabulary; + int max_rule_span; + bool require_tight_phrases; +}; + +#endif diff --git a/extractor/target_phrase_extractor_test.cc b/extractor/target_phrase_extractor_test.cc new file mode 100644 index 00000000..7394f4d9 --- /dev/null +++ b/extractor/target_phrase_extractor_test.cc @@ -0,0 +1,116 @@ +#include + +#include +#include + +#include "mocks/mock_alignment.h" +#include "mocks/mock_data_array.h" +#include "mocks/mock_rule_extractor_helper.h" +#include "mocks/mock_vocabulary.h" +#include "phrase.h" +#include "phrase_builder.h" +#include "target_phrase_extractor.h" + +using namespace std; +using namespace ::testing; + +namespace { + +class TargetPhraseExtractorTest : public Test { + protected: + virtual void SetUp() { + data_array = make_shared(); + alignment = make_shared(); + vocabulary = make_shared(); + phrase_builder = make_shared(vocabulary); + helper = make_shared(); + } + + shared_ptr data_array; + shared_ptr alignment; + shared_ptr vocabulary; + shared_ptr phrase_builder; + shared_ptr helper; + shared_ptr extractor; +}; + +TEST_F(TargetPhraseExtractorTest, TestExtractTightPhrasesTrue) { + EXPECT_CALL(*data_array, GetSentenceLength(1)).WillRepeatedly(Return(5)); + EXPECT_CALL(*data_array, GetSentenceStart(1)).WillRepeatedly(Return(3)); + + vector target_words = {"a", "b", "c", "d", "e"}; + vector target_symbols = {20, 21, 22, 23, 24}; + for (size_t i = 0; i < target_words.size(); ++i) { + EXPECT_CALL(*data_array, GetWordAtIndex(i + 3)) + .WillRepeatedly(Return(target_words[i])); + EXPECT_CALL(*vocabulary, GetTerminalIndex(target_words[i])) + .WillRepeatedly(Return(target_symbols[i])); + EXPECT_CALL(*vocabulary, GetTerminalValue(target_symbols[i])) + .WillRepeatedly(Return(target_words[i])); + } + + vector > links = { + make_pair(0, 0), make_pair(1, 3), make_pair(2, 2), make_pair(3, 1), + make_pair(4, 4) + }; + EXPECT_CALL(*alignment, GetLinks(1)).WillRepeatedly(Return(links)); + + vector gap_order = {1, 0}; + EXPECT_CALL(*helper, GetGapOrder(_)).WillRepeatedly(Return(gap_order)); + + extractor = make_shared( + data_array, alignment, phrase_builder, helper, vocabulary, 10, true); + + vector > target_gaps = {make_pair(3, 4), make_pair(1, 2)}; + vector target_low = {0, 3, 2, 1, 4}; + unordered_map source_indexes = {{0, 0}, {2, 2}, {4, 4}}; + + vector > results = extractor->ExtractPhrases( + target_gaps, target_low, 0, 5, source_indexes, 1); + EXPECT_EQ(1, results.size()); + vector expected_symbols = {20, -2, 22, -1, 24}; + EXPECT_EQ(expected_symbols, results[0].first.Get()); + vector expected_words = {"a", "c", "e"}; + EXPECT_EQ(expected_words, results[0].first.GetWords()); + vector > expected_alignment = { + make_pair(0, 0), make_pair(2, 2), make_pair(4, 4) + }; + EXPECT_EQ(expected_alignment, results[0].second); +} + +TEST_F(TargetPhraseExtractorTest, TestExtractPhrasesTightPhrasesFalse) { + vector target_words = {"a", "b", "c", "d", "e", "f"}; + vector target_symbols = {20, 21, 22, 23, 24, 25, 26}; + EXPECT_CALL(*data_array, GetSentenceLength(0)).WillRepeatedly(Return(6)); + EXPECT_CALL(*data_array, GetSentenceStart(0)).WillRepeatedly(Return(0)); + + for (size_t i = 0; i < target_words.size(); ++i) { + EXPECT_CALL(*data_array, GetWordAtIndex(i)) + .WillRepeatedly(Return(target_words[i])); + EXPECT_CALL(*vocabulary, GetTerminalIndex(target_words[i])) + .WillRepeatedly(Return(target_symbols[i])); + EXPECT_CALL(*vocabulary, GetTerminalValue(target_symbols[i])) + .WillRepeatedly(Return(target_words[i])); + } + + vector > links = {make_pair(1, 1)}; + EXPECT_CALL(*alignment, GetLinks(0)).WillRepeatedly(Return(links)); + + vector gap_order = {0}; + EXPECT_CALL(*helper, GetGapOrder(_)).WillRepeatedly(Return(gap_order)); + + extractor = make_shared( + data_array, alignment, phrase_builder, helper, vocabulary, 10, false); + + vector > target_gaps = {make_pair(2, 4)}; + vector target_low = {-1, 1, -1, -1, -1, -1}; + unordered_map source_indexes = {{1, 1}}; + + vector > results = extractor->ExtractPhrases( + target_gaps, target_low, 1, 5, source_indexes, 0); + EXPECT_EQ(10, results.size()); + // TODO(pauldb): Finish unit test once it's clear how these alignments should + // look like. +} + +} // namespace diff --git a/extractor/translation_table.cc b/extractor/translation_table.cc index 10f1b9ed..a48c0657 100644 --- a/extractor/translation_table.cc +++ b/extractor/translation_table.cc @@ -9,7 +9,6 @@ #include "data_array.h" using namespace std; -using namespace tr1; TranslationTable::TranslationTable(shared_ptr source_data_array, shared_ptr target_data_array, @@ -20,14 +19,15 @@ TranslationTable::TranslationTable(shared_ptr source_data_array, unordered_map source_links_count; unordered_map target_links_count; - unordered_map, int, PairHash > links_count; + unordered_map, int, PairHash> links_count; for (size_t i = 0; i < source_data_array->GetNumSentences(); ++i) { - const vector >& links = alignment->GetLinks(i); + vector > links = alignment->GetLinks(i); int source_start = source_data_array->GetSentenceStart(i); - int next_source_start = source_data_array->GetSentenceStart(i + 1); int target_start = target_data_array->GetSentenceStart(i); - int next_target_start = target_data_array->GetSentenceStart(i + 1); + // Ignore END_OF_LINE markers. + int next_source_start = source_data_array->GetSentenceStart(i + 1) - 1; + int next_target_start = target_data_array->GetSentenceStart(i + 1) - 1; vector source_sentence(source_data.begin() + source_start, source_data.begin() + next_source_start); vector target_sentence(target_data.begin() + target_start, @@ -38,15 +38,23 @@ TranslationTable::TranslationTable(shared_ptr source_data_array, for (pair link: links) { source_linked_words[link.first] = 1; target_linked_words[link.second] = 1; - int source_word = source_sentence[link.first]; - int target_word = target_sentence[link.second]; + IncreaseLinksCount(source_links_count, target_links_count, links_count, + source_sentence[link.first], target_sentence[link.second]); + } - ++source_links_count[source_word]; - ++target_links_count[target_word]; - ++links_count[make_pair(source_word, target_word)]; + for (size_t i = 0; i < source_sentence.size(); ++i) { + if (!source_linked_words[i]) { + IncreaseLinksCount(source_links_count, target_links_count, links_count, + source_sentence[i], DataArray::NULL_WORD); + } } - // TODO(pauldb): Something seems wrong here. No NULL word? + for (size_t i = 0; i < target_sentence.size(); ++i) { + if (!target_linked_words[i]) { + IncreaseLinksCount(source_links_count, target_links_count, links_count, + DataArray::NULL_WORD, target_sentence[i]); + } + } } for (pair, int> link_count: links_count) { @@ -58,6 +66,21 @@ TranslationTable::TranslationTable(shared_ptr source_data_array, } } +TranslationTable::TranslationTable() {} + +TranslationTable::~TranslationTable() {} + +void TranslationTable::IncreaseLinksCount( + unordered_map& source_links_count, + unordered_map& target_links_count, + unordered_map, int, PairHash>& links_count, + int source_word_id, + int target_word_id) const { + ++source_links_count[source_word_id]; + ++target_links_count[target_word_id]; + ++links_count[make_pair(source_word_id, target_word_id)]; +} + double TranslationTable::GetTargetGivenSourceScore( const string& source_word, const string& target_word) { if (!source_data_array->HasWord(source_word) || @@ -73,7 +96,7 @@ double TranslationTable::GetTargetGivenSourceScore( double TranslationTable::GetSourceGivenTargetScore( const string& source_word, const string& target_word) { if (!source_data_array->HasWord(source_word) || - !target_data_array->HasWord(target_word) == 0) { + !target_data_array->HasWord(target_word)) { return -1; } diff --git a/extractor/translation_table.h b/extractor/translation_table.h index acf94af7..157ad3af 100644 --- a/extractor/translation_table.h +++ b/extractor/translation_table.h @@ -3,13 +3,12 @@ #include #include -#include +#include #include #include using namespace std; -using namespace tr1; namespace fs = boost::filesystem; class Alignment; @@ -24,15 +23,27 @@ class TranslationTable { shared_ptr target_data_array, shared_ptr alignment); - double GetTargetGivenSourceScore(const string& source_word, - const string& target_word); + virtual ~TranslationTable(); - double GetSourceGivenTargetScore(const string& source_word, - const string& target_word); + virtual double GetTargetGivenSourceScore(const string& source_word, + const string& target_word); + + virtual double GetSourceGivenTargetScore(const string& source_word, + const string& target_word); void WriteBinary(const fs::path& filepath) const; + protected: + TranslationTable(); + private: + void IncreaseLinksCount( + unordered_map& source_links_count, + unordered_map& target_links_count, + unordered_map, int, PairHash>& links_count, + int source_word_id, + int target_word_id) const; + shared_ptr source_data_array; shared_ptr target_data_array; unordered_map, pair, PairHash> diff --git a/extractor/translation_table_test.cc b/extractor/translation_table_test.cc new file mode 100644 index 00000000..c99f3f93 --- /dev/null +++ b/extractor/translation_table_test.cc @@ -0,0 +1,82 @@ +#include + +#include +#include +#include + +#include "mocks/mock_alignment.h" +#include "mocks/mock_data_array.h" +#include "translation_table.h" + +using namespace std; +using namespace ::testing; + +namespace { + +TEST(TranslationTableTest, TestScores) { + vector words = {"a", "b", "c"}; + + vector source_data = {2, 3, 2, 3, 4, 0, 2, 3, 6, 0, 2, 3, 6, 0}; + vector source_sentence_start = {0, 6, 10, 14}; + shared_ptr source_data_array = make_shared(); + EXPECT_CALL(*source_data_array, GetData()) + .WillRepeatedly(ReturnRef(source_data)); + EXPECT_CALL(*source_data_array, GetNumSentences()) + .WillRepeatedly(Return(3)); + for (size_t i = 0; i < source_sentence_start.size(); ++i) { + EXPECT_CALL(*source_data_array, GetSentenceStart(i)) + .WillRepeatedly(Return(source_sentence_start[i])); + } + for (size_t i = 0; i < words.size(); ++i) { + EXPECT_CALL(*source_data_array, HasWord(words[i])) + .WillRepeatedly(Return(true)); + EXPECT_CALL(*source_data_array, GetWordId(words[i])) + .WillRepeatedly(Return(i + 2)); + } + EXPECT_CALL(*source_data_array, HasWord("d")) + .WillRepeatedly(Return(false)); + + vector target_data = {2, 3, 2, 3, 4, 5, 0, 3, 6, 0, 2, 7, 0}; + vector target_sentence_start = {0, 7, 10, 13}; + shared_ptr target_data_array = make_shared(); + EXPECT_CALL(*target_data_array, GetData()) + .WillRepeatedly(ReturnRef(target_data)); + for (size_t i = 0; i < target_sentence_start.size(); ++i) { + EXPECT_CALL(*target_data_array, GetSentenceStart(i)) + .WillRepeatedly(Return(target_sentence_start[i])); + } + for (size_t i = 0; i < words.size(); ++i) { + EXPECT_CALL(*target_data_array, HasWord(words[i])) + .WillRepeatedly(Return(true)); + EXPECT_CALL(*target_data_array, GetWordId(words[i])) + .WillRepeatedly(Return(i + 2)); + } + EXPECT_CALL(*target_data_array, HasWord("d")) + .WillRepeatedly(Return(false)); + + vector > links1 = { + make_pair(0, 0), make_pair(1, 1), make_pair(2, 2), make_pair(3, 3), + make_pair(4, 4), make_pair(4, 5) + }; + vector > links2 = {make_pair(1, 0), make_pair(2, 1)}; + vector > links3 = {make_pair(0, 0), make_pair(2, 1)}; + shared_ptr alignment = make_shared(); + EXPECT_CALL(*alignment, GetLinks(0)).WillRepeatedly(Return(links1)); + EXPECT_CALL(*alignment, GetLinks(1)).WillRepeatedly(Return(links2)); + EXPECT_CALL(*alignment, GetLinks(2)).WillRepeatedly(Return(links3)); + + shared_ptr table = make_shared( + source_data_array, target_data_array, alignment); + + EXPECT_EQ(0.75, table->GetTargetGivenSourceScore("a", "a")); + EXPECT_EQ(0, table->GetTargetGivenSourceScore("a", "b")); + EXPECT_EQ(0.5, table->GetTargetGivenSourceScore("c", "c")); + EXPECT_EQ(-1, table->GetTargetGivenSourceScore("c", "d")); + + EXPECT_EQ(1, table->GetSourceGivenTargetScore("a", "a")); + EXPECT_EQ(0, table->GetSourceGivenTargetScore("a", "b")); + EXPECT_EQ(1, table->GetSourceGivenTargetScore("c", "c")); + EXPECT_EQ(-1, table->GetSourceGivenTargetScore("c", "d")); +} + +} // namespace diff --git a/extractor/vocabulary.h b/extractor/vocabulary.h index ed55e5e4..c6a8b3e8 100644 --- a/extractor/vocabulary.h +++ b/extractor/vocabulary.h @@ -2,11 +2,10 @@ #define _VOCABULARY_H_ #include -#include +#include #include using namespace std; -using namespace tr1; class Vocabulary { public: -- cgit v1.2.3 From e1ffc4886b98f7ddcd2ec30d740aa2de8282cd8e Mon Sep 17 00:00:00 2001 From: Paul Baltescu Date: Wed, 6 Mar 2013 15:25:54 +0000 Subject: Namespace for extractor. --- extractor/alignment.cc | 4 ++++ extractor/alignment.h | 4 ++++ extractor/alignment_test.cc | 4 +++- extractor/compile.cc | 1 + extractor/data_array.cc | 4 ++++ extractor/data_array.h | 4 ++++ extractor/data_array_test.cc | 4 +++- extractor/fast_intersector.cc | 4 ++++ extractor/fast_intersector.h | 4 ++++ extractor/fast_intersector_test.cc | 8 ++++---- extractor/features/count_source_target.cc | 6 ++++++ extractor/features/count_source_target.h | 6 ++++++ extractor/features/count_source_target_test.cc | 6 +++++- extractor/features/feature.cc | 6 ++++++ extractor/features/feature.h | 6 ++++++ extractor/features/is_source_singleton.cc | 6 ++++++ extractor/features/is_source_singleton.h | 6 ++++++ extractor/features/is_source_singleton_test.cc | 6 +++++- extractor/features/is_source_target_singleton.cc | 6 ++++++ extractor/features/is_source_target_singleton.h | 6 ++++++ extractor/features/is_source_target_singleton_test.cc | 6 +++++- extractor/features/max_lex_source_given_target.cc | 6 ++++++ extractor/features/max_lex_source_given_target.h | 7 +++++++ extractor/features/max_lex_source_given_target_test.cc | 6 +++++- extractor/features/max_lex_target_given_source.cc | 6 ++++++ extractor/features/max_lex_target_given_source.h | 7 +++++++ extractor/features/max_lex_target_given_source_test.cc | 6 +++++- extractor/features/sample_source_count.cc | 6 ++++++ extractor/features/sample_source_count.h | 6 ++++++ extractor/features/sample_source_count_test.cc | 6 +++++- extractor/features/target_given_source_coherent.cc | 6 ++++++ extractor/features/target_given_source_coherent.h | 6 ++++++ extractor/features/target_given_source_coherent_test.cc | 6 +++++- extractor/grammar.cc | 4 ++++ extractor/grammar.h | 4 ++++ extractor/grammar_extractor.cc | 5 +++++ extractor/grammar_extractor.h | 9 +++++++-- extractor/grammar_extractor_test.cc | 4 +++- extractor/matchings_finder.cc | 4 ++++ extractor/matchings_finder.h | 4 ++++ extractor/matchings_finder_test.cc | 4 +++- extractor/matchings_trie.cc | 4 ++++ extractor/matchings_trie.h | 4 ++++ extractor/mocks/mock_alignment.h | 4 ++++ extractor/mocks/mock_data_array.h | 4 ++++ extractor/mocks/mock_fast_intersector.h | 4 ++++ extractor/mocks/mock_feature.h | 6 ++++++ extractor/mocks/mock_matchings_finder.h | 4 ++++ extractor/mocks/mock_precomputation.h | 4 ++++ extractor/mocks/mock_rule_extractor.h | 4 ++++ extractor/mocks/mock_rule_extractor_helper.h | 4 ++++ extractor/mocks/mock_rule_factory.h | 4 ++++ extractor/mocks/mock_sampler.h | 4 ++++ extractor/mocks/mock_scorer.h | 7 ++++++- extractor/mocks/mock_suffix_array.h | 4 ++++ extractor/mocks/mock_target_phrase_extractor.h | 4 ++++ extractor/mocks/mock_translation_table.h | 4 ++++ extractor/mocks/mock_vocabulary.h | 4 ++++ extractor/phrase.cc | 4 ++++ extractor/phrase.h | 4 ++++ extractor/phrase_builder.cc | 4 ++++ extractor/phrase_builder.h | 4 ++++ extractor/phrase_location.cc | 4 ++++ extractor/phrase_location.h | 4 ++++ extractor/phrase_test.cc | 4 +++- extractor/precomputation.cc | 4 ++++ extractor/precomputation.h | 6 +++++- extractor/precomputation_test.cc | 5 ++++- extractor/rule.cc | 4 ++++ extractor/rule.h | 4 ++++ extractor/rule_extractor.cc | 6 +++++- extractor/rule_extractor.h | 8 ++++++-- extractor/rule_extractor_helper.cc | 4 ++++ extractor/rule_extractor_helper.h | 4 ++++ extractor/rule_extractor_helper_test.cc | 4 +++- extractor/rule_extractor_test.cc | 4 +++- extractor/rule_factory.cc | 5 +++++ extractor/rule_factory.h | 8 ++++++-- extractor/rule_factory_test.cc | 4 +++- extractor/run_extractor.cc | 17 ++++++++++------- extractor/sampler.cc | 4 ++++ extractor/sampler.h | 4 ++++ extractor/sampler_test.cc | 4 +++- extractor/scorer.cc | 8 ++++++-- extractor/scorer.h | 16 +++++++++++----- extractor/scorer_test.cc | 16 +++++++++------- extractor/suffix_array.cc | 4 ++++ extractor/suffix_array.h | 4 ++++ extractor/suffix_array_test.cc | 4 +++- extractor/target_phrase_extractor.cc | 4 ++++ extractor/target_phrase_extractor.h | 8 ++++++-- extractor/target_phrase_extractor_test.cc | 4 +++- extractor/time_util.cc | 4 ++++ extractor/time_util.h | 4 ++++ extractor/translation_table.cc | 4 ++++ extractor/translation_table.h | 8 ++++++-- extractor/translation_table_test.cc | 4 +++- extractor/vocabulary.cc | 3 +++ extractor/vocabulary.h | 4 ++++ 99 files changed, 460 insertions(+), 58 deletions(-) (limited to 'extractor/features/max_lex_target_given_source.cc') diff --git a/extractor/alignment.cc b/extractor/alignment.cc index ff39d484..f9bbcf6a 100644 --- a/extractor/alignment.cc +++ b/extractor/alignment.cc @@ -13,6 +13,8 @@ namespace fs = boost::filesystem; using namespace std; +namespace extractor { + Alignment::Alignment(const string& filename) { ifstream infile(filename.c_str()); string line; @@ -49,3 +51,5 @@ void Alignment::WriteBinary(const fs::path& filepath) { fwrite(alignment.data(), sizeof(pair), size, file); } } + +} // namespace extractor diff --git a/extractor/alignment.h b/extractor/alignment.h index f7e79585..ef89dc0c 100644 --- a/extractor/alignment.h +++ b/extractor/alignment.h @@ -9,6 +9,8 @@ namespace fs = boost::filesystem; using namespace std; +namespace extractor { + class Alignment { public: Alignment(const string& filename); @@ -26,4 +28,6 @@ class Alignment { vector > > alignments; }; +} // namespace extractor + #endif diff --git a/extractor/alignment_test.cc b/extractor/alignment_test.cc index 1bc51a56..a7defb66 100644 --- a/extractor/alignment_test.cc +++ b/extractor/alignment_test.cc @@ -8,6 +8,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class AlignmentTest : public Test { @@ -28,4 +29,5 @@ TEST_F(AlignmentTest, TestGetLinks) { EXPECT_EQ(expected_links, alignment->GetLinks(1)); } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/compile.cc b/extractor/compile.cc index f5cd41f4..7062ef03 100644 --- a/extractor/compile.cc +++ b/extractor/compile.cc @@ -14,6 +14,7 @@ namespace fs = boost::filesystem; namespace po = boost::program_options; using namespace std; +using namespace extractor; int main(int argc, char** argv) { po::options_description desc("Command line options"); diff --git a/extractor/data_array.cc b/extractor/data_array.cc index cd430c69..481abb80 100644 --- a/extractor/data_array.cc +++ b/extractor/data_array.cc @@ -10,6 +10,8 @@ namespace fs = boost::filesystem; using namespace std; +namespace extractor { + int DataArray::NULL_WORD = 0; int DataArray::END_OF_LINE = 1; string DataArray::NULL_WORD_STR = "__NULL__"; @@ -154,3 +156,5 @@ int DataArray::GetWordId(const string& word) const { string DataArray::GetWord(int word_id) const { return id2word[word_id]; } + +} // namespace extractor diff --git a/extractor/data_array.h b/extractor/data_array.h index 96950789..42e12135 100644 --- a/extractor/data_array.h +++ b/extractor/data_array.h @@ -10,6 +10,8 @@ namespace fs = boost::filesystem; using namespace std; +namespace extractor { + enum Side { SOURCE, TARGET @@ -73,4 +75,6 @@ class DataArray { vector sentence_start; }; +} // namespace extractor + #endif diff --git a/extractor/data_array_test.cc b/extractor/data_array_test.cc index ba5ce09e..71175fda 100644 --- a/extractor/data_array_test.cc +++ b/extractor/data_array_test.cc @@ -11,6 +11,7 @@ using namespace std; using namespace ::testing; namespace fs = boost::filesystem; +namespace extractor { namespace { class DataArrayTest : public Test { @@ -93,4 +94,5 @@ TEST_F(DataArrayTest, TestSentenceData) { } } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/fast_intersector.cc b/extractor/fast_intersector.cc index 8c7a7af8..cec3d30b 100644 --- a/extractor/fast_intersector.cc +++ b/extractor/fast_intersector.cc @@ -9,6 +9,8 @@ #include "suffix_array.h" #include "vocabulary.h" +namespace extractor { + FastIntersector::FastIntersector(shared_ptr suffix_array, shared_ptr precomputation, shared_ptr vocabulary, @@ -189,3 +191,5 @@ pair FastIntersector::GetSearchRange(bool has_marginal_x) const { return make_pair(1, 2); } } + +} // namespace extractor diff --git a/extractor/fast_intersector.h b/extractor/fast_intersector.h index 785e428e..32c88a30 100644 --- a/extractor/fast_intersector.h +++ b/extractor/fast_intersector.h @@ -9,6 +9,8 @@ using namespace std; +namespace extractor { + typedef boost::hash > VectorHash; typedef unordered_map, vector, VectorHash> Index; @@ -62,4 +64,6 @@ class FastIntersector { Index collocations; }; +} // namespace extractor + #endif diff --git a/extractor/fast_intersector_test.cc b/extractor/fast_intersector_test.cc index 0d6ef367..76c3aaea 100644 --- a/extractor/fast_intersector_test.cc +++ b/extractor/fast_intersector_test.cc @@ -4,8 +4,8 @@ #include "fast_intersector.h" #include "mocks/mock_data_array.h" -#include "mocks/mock_suffix_array.h" #include "mocks/mock_precomputation.h" +#include "mocks/mock_suffix_array.h" #include "mocks/mock_vocabulary.h" #include "phrase.h" #include "phrase_location.h" @@ -14,6 +14,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class FastIntersectorTest : public Test { @@ -112,7 +113,6 @@ TEST_F(FastIntersectorTest, TestIntersectaXbXcExtendSuffix) { EXPECT_EQ(PhraseLocation(expected_locs, 3), result); } -/* TEST_F(FastIntersectorTest, TestIntersectaXbExtendPrefix) { vector symbols = {1, -1, 3}; Phrase phrase = phrase_builder->Build(symbols); @@ -141,6 +141,6 @@ TEST_F(FastIntersectorTest, TestIntersectCheckEstimates) { EXPECT_EQ(PhraseLocation(expected_locs, 2), result); EXPECT_EQ(PhraseLocation(10, 12), suffix_location); } -*/ -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/features/count_source_target.cc b/extractor/features/count_source_target.cc index 9441b451..db0385e0 100644 --- a/extractor/features/count_source_target.cc +++ b/extractor/features/count_source_target.cc @@ -2,6 +2,9 @@ #include +namespace extractor { +namespace features { + double CountSourceTarget::Score(const FeatureContext& context) const { return log10(1 + context.pair_count); } @@ -9,3 +12,6 @@ double CountSourceTarget::Score(const FeatureContext& context) const { string CountSourceTarget::GetName() const { return "CountEF"; } + +} // namespace features +} // namespace extractor diff --git a/extractor/features/count_source_target.h b/extractor/features/count_source_target.h index a2481944..dec78883 100644 --- a/extractor/features/count_source_target.h +++ b/extractor/features/count_source_target.h @@ -3,6 +3,9 @@ #include "feature.h" +namespace extractor { +namespace features { + class CountSourceTarget : public Feature { public: double Score(const FeatureContext& context) const; @@ -10,4 +13,7 @@ class CountSourceTarget : public Feature { string GetName() const; }; +} // namespace features +} // namespace extractor + #endif diff --git a/extractor/features/count_source_target_test.cc b/extractor/features/count_source_target_test.cc index 22633bb6..1fd0c2aa 100644 --- a/extractor/features/count_source_target_test.cc +++ b/extractor/features/count_source_target_test.cc @@ -8,6 +8,8 @@ using namespace std; using namespace ::testing; +namespace extractor { +namespace features { namespace { class CountSourceTargetTest : public Test { @@ -29,4 +31,6 @@ TEST_F(CountSourceTargetTest, TestScore) { EXPECT_EQ(1.0, feature->Score(context)); } -} // namespace +} // namespace +} // namespace features +} // namespace extractor diff --git a/extractor/features/feature.cc b/extractor/features/feature.cc index 876f5f8f..939bcc59 100644 --- a/extractor/features/feature.cc +++ b/extractor/features/feature.cc @@ -1,5 +1,11 @@ #include "feature.h" +namespace extractor { +namespace features { + const double Feature::MAX_SCORE = 99.0; Feature::~Feature() {} + +} // namespace features +} // namespace extractor diff --git a/extractor/features/feature.h b/extractor/features/feature.h index aca58401..de2827bc 100644 --- a/extractor/features/feature.h +++ b/extractor/features/feature.h @@ -8,6 +8,9 @@ using namespace std; +namespace extractor { +namespace features { + struct FeatureContext { FeatureContext(const Phrase& source_phrase, const Phrase& target_phrase, double source_phrase_count, int pair_count, int num_samples) : @@ -33,4 +36,7 @@ class Feature { static const double MAX_SCORE; }; +} // namespace features +} // namespace extractor + #endif diff --git a/extractor/features/is_source_singleton.cc b/extractor/features/is_source_singleton.cc index 98d4e5fe..ab54e51a 100644 --- a/extractor/features/is_source_singleton.cc +++ b/extractor/features/is_source_singleton.cc @@ -2,6 +2,9 @@ #include +namespace extractor { +namespace features { + double IsSourceSingleton::Score(const FeatureContext& context) const { return context.source_phrase_count == 1; } @@ -9,3 +12,6 @@ double IsSourceSingleton::Score(const FeatureContext& context) const { string IsSourceSingleton::GetName() const { return "IsSingletonF"; } + +} // namespace features +} // namespace extractor diff --git a/extractor/features/is_source_singleton.h b/extractor/features/is_source_singleton.h index 7cc72828..30f76c6d 100644 --- a/extractor/features/is_source_singleton.h +++ b/extractor/features/is_source_singleton.h @@ -3,6 +3,9 @@ #include "feature.h" +namespace extractor { +namespace features { + class IsSourceSingleton : public Feature { public: double Score(const FeatureContext& context) const; @@ -10,4 +13,7 @@ class IsSourceSingleton : public Feature { string GetName() const; }; +} // namespace features +} // namespace extractor + #endif diff --git a/extractor/features/is_source_singleton_test.cc b/extractor/features/is_source_singleton_test.cc index 8c71e593..f4266671 100644 --- a/extractor/features/is_source_singleton_test.cc +++ b/extractor/features/is_source_singleton_test.cc @@ -8,6 +8,8 @@ using namespace std; using namespace ::testing; +namespace extractor { +namespace features { namespace { class IsSourceSingletonTest : public Test { @@ -32,4 +34,6 @@ TEST_F(IsSourceSingletonTest, TestScore) { EXPECT_EQ(1, feature->Score(context)); } -} // namespace +} // namespace +} // namespace features +} // namespace extractor diff --git a/extractor/features/is_source_target_singleton.cc b/extractor/features/is_source_target_singleton.cc index 31d36532..03b3c62c 100644 --- a/extractor/features/is_source_target_singleton.cc +++ b/extractor/features/is_source_target_singleton.cc @@ -2,6 +2,9 @@ #include +namespace extractor { +namespace features { + double IsSourceTargetSingleton::Score(const FeatureContext& context) const { return context.pair_count == 1; } @@ -9,3 +12,6 @@ double IsSourceTargetSingleton::Score(const FeatureContext& context) const { string IsSourceTargetSingleton::GetName() const { return "IsSingletonFE"; } + +} // namespace features +} // namespace extractor diff --git a/extractor/features/is_source_target_singleton.h b/extractor/features/is_source_target_singleton.h index 58913b74..12fb6ee6 100644 --- a/extractor/features/is_source_target_singleton.h +++ b/extractor/features/is_source_target_singleton.h @@ -3,6 +3,9 @@ #include "feature.h" +namespace extractor { +namespace features { + class IsSourceTargetSingleton : public Feature { public: double Score(const FeatureContext& context) const; @@ -10,4 +13,7 @@ class IsSourceTargetSingleton : public Feature { string GetName() const; }; +} // namespace features +} // namespace extractor + #endif diff --git a/extractor/features/is_source_target_singleton_test.cc b/extractor/features/is_source_target_singleton_test.cc index a51f77c9..929635b0 100644 --- a/extractor/features/is_source_target_singleton_test.cc +++ b/extractor/features/is_source_target_singleton_test.cc @@ -8,6 +8,8 @@ using namespace std; using namespace ::testing; +namespace extractor { +namespace features { namespace { class IsSourceTargetSingletonTest : public Test { @@ -32,4 +34,6 @@ TEST_F(IsSourceTargetSingletonTest, TestScore) { EXPECT_EQ(1, feature->Score(context)); } -} // namespace +} // namespace +} // namespace features +} // namespace extractor diff --git a/extractor/features/max_lex_source_given_target.cc b/extractor/features/max_lex_source_given_target.cc index 21f5c76a..3ffe598c 100644 --- a/extractor/features/max_lex_source_given_target.cc +++ b/extractor/features/max_lex_source_given_target.cc @@ -5,6 +5,9 @@ #include "../data_array.h" #include "../translation_table.h" +namespace extractor { +namespace features { + MaxLexSourceGivenTarget::MaxLexSourceGivenTarget( shared_ptr table) : table(table) {} @@ -29,3 +32,6 @@ double MaxLexSourceGivenTarget::Score(const FeatureContext& context) const { string MaxLexSourceGivenTarget::GetName() const { return "MaxLexFgivenE"; } + +} // namespace features +} // namespace extractor diff --git a/extractor/features/max_lex_source_given_target.h b/extractor/features/max_lex_source_given_target.h index e87c1c8e..bfa7ef1b 100644 --- a/extractor/features/max_lex_source_given_target.h +++ b/extractor/features/max_lex_source_given_target.h @@ -7,8 +7,12 @@ using namespace std; +namespace extractor { + class TranslationTable; +namespace features { + class MaxLexSourceGivenTarget : public Feature { public: MaxLexSourceGivenTarget(shared_ptr table); @@ -21,4 +25,7 @@ class MaxLexSourceGivenTarget : public Feature { shared_ptr table; }; +} // namespace features +} // namespace extractor + #endif diff --git a/extractor/features/max_lex_source_given_target_test.cc b/extractor/features/max_lex_source_given_target_test.cc index 5fd41f8b..c1edb483 100644 --- a/extractor/features/max_lex_source_given_target_test.cc +++ b/extractor/features/max_lex_source_given_target_test.cc @@ -13,6 +13,8 @@ using namespace std; using namespace ::testing; +namespace extractor { +namespace features { namespace { class MaxLexSourceGivenTargetTest : public Test { @@ -71,4 +73,6 @@ TEST_F(MaxLexSourceGivenTargetTest, TestScore) { EXPECT_EQ(99 - log10(18), feature->Score(context)); } -} // namespace +} // namespace +} // namespace features +} // namespace extractor diff --git a/extractor/features/max_lex_target_given_source.cc b/extractor/features/max_lex_target_given_source.cc index f2bc2474..30140d80 100644 --- a/extractor/features/max_lex_target_given_source.cc +++ b/extractor/features/max_lex_target_given_source.cc @@ -5,6 +5,9 @@ #include "../data_array.h" #include "../translation_table.h" +namespace extractor { +namespace features { + MaxLexTargetGivenSource::MaxLexTargetGivenSource( shared_ptr table) : table(table) {} @@ -29,3 +32,6 @@ double MaxLexTargetGivenSource::Score(const FeatureContext& context) const { string MaxLexTargetGivenSource::GetName() const { return "MaxLexEgivenF"; } + +} // namespace features +} // namespace extractor diff --git a/extractor/features/max_lex_target_given_source.h b/extractor/features/max_lex_target_given_source.h index 9585ff04..66cf0914 100644 --- a/extractor/features/max_lex_target_given_source.h +++ b/extractor/features/max_lex_target_given_source.h @@ -7,8 +7,12 @@ using namespace std; +namespace extractor { + class TranslationTable; +namespace features { + class MaxLexTargetGivenSource : public Feature { public: MaxLexTargetGivenSource(shared_ptr table); @@ -21,4 +25,7 @@ class MaxLexTargetGivenSource : public Feature { shared_ptr table; }; +} // namespace features +} // namespace extractor + #endif diff --git a/extractor/features/max_lex_target_given_source_test.cc b/extractor/features/max_lex_target_given_source_test.cc index c8701bf7..9ceb13e5 100644 --- a/extractor/features/max_lex_target_given_source_test.cc +++ b/extractor/features/max_lex_target_given_source_test.cc @@ -13,6 +13,8 @@ using namespace std; using namespace ::testing; +namespace extractor { +namespace features { namespace { class MaxLexTargetGivenSourceTest : public Test { @@ -71,4 +73,6 @@ TEST_F(MaxLexTargetGivenSourceTest, TestScore) { EXPECT_EQ(-log10(36), feature->Score(context)); } -} // namespace +} // namespace +} // namespace features +} // namespace extractor diff --git a/extractor/features/sample_source_count.cc b/extractor/features/sample_source_count.cc index 88b645b1..b110fc51 100644 --- a/extractor/features/sample_source_count.cc +++ b/extractor/features/sample_source_count.cc @@ -2,6 +2,9 @@ #include +namespace extractor { +namespace features { + double SampleSourceCount::Score(const FeatureContext& context) const { return log10(1 + context.num_samples); } @@ -9,3 +12,6 @@ double SampleSourceCount::Score(const FeatureContext& context) const { string SampleSourceCount::GetName() const { return "SampleCountF"; } + +} // namespace features +} // namespace extractor diff --git a/extractor/features/sample_source_count.h b/extractor/features/sample_source_count.h index 62d236c8..53c7f954 100644 --- a/extractor/features/sample_source_count.h +++ b/extractor/features/sample_source_count.h @@ -3,6 +3,9 @@ #include "feature.h" +namespace extractor { +namespace features { + class SampleSourceCount : public Feature { public: double Score(const FeatureContext& context) const; @@ -10,4 +13,7 @@ class SampleSourceCount : public Feature { string GetName() const; }; +} // namespace features +} // namespace extractor + #endif diff --git a/extractor/features/sample_source_count_test.cc b/extractor/features/sample_source_count_test.cc index 7d226104..63856b9d 100644 --- a/extractor/features/sample_source_count_test.cc +++ b/extractor/features/sample_source_count_test.cc @@ -9,6 +9,8 @@ using namespace std; using namespace ::testing; +namespace extractor { +namespace features { namespace { class SampleSourceCountTest : public Test { @@ -33,4 +35,6 @@ TEST_F(SampleSourceCountTest, TestScore) { EXPECT_EQ(1.0, feature->Score(context)); } -} // namespace +} // namespace +} // namespace features +} // namespace extractor diff --git a/extractor/features/target_given_source_coherent.cc b/extractor/features/target_given_source_coherent.cc index 274b3364..c4551d88 100644 --- a/extractor/features/target_given_source_coherent.cc +++ b/extractor/features/target_given_source_coherent.cc @@ -2,6 +2,9 @@ #include +namespace extractor { +namespace features { + double TargetGivenSourceCoherent::Score(const FeatureContext& context) const { double prob = (double) context.pair_count / context.num_samples; return prob > 0 ? -log10(prob) : MAX_SCORE; @@ -10,3 +13,6 @@ double TargetGivenSourceCoherent::Score(const FeatureContext& context) const { string TargetGivenSourceCoherent::GetName() const { return "EgivenFCoherent"; } + +} // namespace features +} // namespace extractor diff --git a/extractor/features/target_given_source_coherent.h b/extractor/features/target_given_source_coherent.h index 09c8edb1..80d9f617 100644 --- a/extractor/features/target_given_source_coherent.h +++ b/extractor/features/target_given_source_coherent.h @@ -3,6 +3,9 @@ #include "feature.h" +namespace extractor { +namespace features { + class TargetGivenSourceCoherent : public Feature { public: double Score(const FeatureContext& context) const; @@ -10,4 +13,7 @@ class TargetGivenSourceCoherent : public Feature { string GetName() const; }; +} // namespace features +} // namespace extractor + #endif diff --git a/extractor/features/target_given_source_coherent_test.cc b/extractor/features/target_given_source_coherent_test.cc index c54c06c2..454105e1 100644 --- a/extractor/features/target_given_source_coherent_test.cc +++ b/extractor/features/target_given_source_coherent_test.cc @@ -8,6 +8,8 @@ using namespace std; using namespace ::testing; +namespace extractor { +namespace features { namespace { class TargetGivenSourceCoherentTest : public Test { @@ -32,4 +34,6 @@ TEST_F(TargetGivenSourceCoherentTest, TestScore) { EXPECT_EQ(99.0, feature->Score(context)); } -} // namespace +} // namespace +} // namespace features +} // namespace extractor diff --git a/extractor/grammar.cc b/extractor/grammar.cc index 8124a804..8e5bcd45 100644 --- a/extractor/grammar.cc +++ b/extractor/grammar.cc @@ -6,6 +6,8 @@ using namespace std; +namespace extractor { + Grammar::Grammar(const vector& rules, const vector& feature_names) : rules(rules), feature_names(feature_names) {} @@ -37,3 +39,5 @@ ostream& operator<<(ostream& os, const Grammar& grammar) { return os; } + +} // namespace extractor diff --git a/extractor/grammar.h b/extractor/grammar.h index 889cc2f3..a424d65a 100644 --- a/extractor/grammar.h +++ b/extractor/grammar.h @@ -7,6 +7,8 @@ using namespace std; +namespace extractor { + class Rule; class Grammar { @@ -24,4 +26,6 @@ class Grammar { vector feature_names; }; +} // namespace extractor + #endif diff --git a/extractor/grammar_extractor.cc b/extractor/grammar_extractor.cc index b8f6f0c7..8050ce7b 100644 --- a/extractor/grammar_extractor.cc +++ b/extractor/grammar_extractor.cc @@ -6,10 +6,13 @@ #include "grammar.h" #include "rule.h" +#include "rule_factory.h" #include "vocabulary.h" using namespace std; +namespace extractor { + GrammarExtractor::GrammarExtractor( shared_ptr source_suffix_array, shared_ptr target_data_array, @@ -55,3 +58,5 @@ vector GrammarExtractor::AnnotateWords(const vector& words) { } return result; } + +} // namespace extractor diff --git a/extractor/grammar_extractor.h b/extractor/grammar_extractor.h index f50a8d14..6b1dcf98 100644 --- a/extractor/grammar_extractor.h +++ b/extractor/grammar_extractor.h @@ -1,18 +1,21 @@ #ifndef _GRAMMAR_EXTRACTOR_H_ #define _GRAMMAR_EXTRACTOR_H_ +#include #include #include -#include "rule_factory.h" - using namespace std; +namespace extractor { + class Alignment; class DataArray; class Grammar; +class HieroCachingRuleFactory; class Precomputation; class Rule; +class Scorer; class SuffixArray; class Vocabulary; @@ -46,4 +49,6 @@ class GrammarExtractor { shared_ptr rule_factory; }; +} // namespace extractor + #endif diff --git a/extractor/grammar_extractor_test.cc b/extractor/grammar_extractor_test.cc index d4ed7d4f..823bb8b4 100644 --- a/extractor/grammar_extractor_test.cc +++ b/extractor/grammar_extractor_test.cc @@ -13,6 +13,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { TEST(GrammarExtractorTest, TestAnnotatingWords) { @@ -46,4 +47,5 @@ TEST(GrammarExtractorTest, TestAnnotatingWords) { extractor.GetGrammar(sentence); } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/matchings_finder.cc b/extractor/matchings_finder.cc index eaf493b2..ceed6891 100644 --- a/extractor/matchings_finder.cc +++ b/extractor/matchings_finder.cc @@ -3,6 +3,8 @@ #include "suffix_array.h" #include "phrase_location.h" +namespace extractor { + MatchingsFinder::MatchingsFinder(shared_ptr suffix_array) : suffix_array(suffix_array) {} @@ -19,3 +21,5 @@ PhraseLocation MatchingsFinder::Find(PhraseLocation& location, return suffix_array->Lookup(location.sa_low, location.sa_high, word, offset); } + +} // namespace extractor diff --git a/extractor/matchings_finder.h b/extractor/matchings_finder.h index ed04d8b8..fbb504ef 100644 --- a/extractor/matchings_finder.h +++ b/extractor/matchings_finder.h @@ -6,6 +6,8 @@ using namespace std; +namespace extractor { + class PhraseLocation; class SuffixArray; @@ -25,4 +27,6 @@ class MatchingsFinder { shared_ptr suffix_array; }; +} // namespace extractor + #endif diff --git a/extractor/matchings_finder_test.cc b/extractor/matchings_finder_test.cc index 817f1635..d40e5191 100644 --- a/extractor/matchings_finder_test.cc +++ b/extractor/matchings_finder_test.cc @@ -9,6 +9,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class MatchingsFinderTest : public Test { @@ -39,4 +40,5 @@ TEST_F(MatchingsFinderTest, ResizeUnsetRange) { EXPECT_EQ(PhraseLocation(0, 10), phrase_location); } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/matchings_trie.cc b/extractor/matchings_trie.cc index 8ea795db..c7b98765 100644 --- a/extractor/matchings_trie.cc +++ b/extractor/matchings_trie.cc @@ -1,5 +1,7 @@ #include "matchings_trie.h" +namespace extractor { + void MatchingsTrie::Reset() { ResetTree(root); root = make_shared(); @@ -20,3 +22,5 @@ void MatchingsTrie::ResetTree(shared_ptr root) { root.reset(); } } + +} // namespace extractor diff --git a/extractor/matchings_trie.h b/extractor/matchings_trie.h index 6e72b2db..a54671d2 100644 --- a/extractor/matchings_trie.h +++ b/extractor/matchings_trie.h @@ -9,6 +9,8 @@ using namespace std; +namespace extractor { + struct TrieNode { TrieNode(shared_ptr suffix_link = shared_ptr(), Phrase phrase = Phrase(), @@ -44,4 +46,6 @@ class MatchingsTrie { shared_ptr root; }; +} // namespace extractor + #endif diff --git a/extractor/mocks/mock_alignment.h b/extractor/mocks/mock_alignment.h index 4a5077ad..3d745e9d 100644 --- a/extractor/mocks/mock_alignment.h +++ b/extractor/mocks/mock_alignment.h @@ -2,9 +2,13 @@ #include "../alignment.h" +namespace extractor { + typedef vector > SentenceLinks; class MockAlignment : public Alignment { public: MOCK_CONST_METHOD1(GetLinks, SentenceLinks(int sentence_id)); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_data_array.h b/extractor/mocks/mock_data_array.h index 004e8906..cf9f3671 100644 --- a/extractor/mocks/mock_data_array.h +++ b/extractor/mocks/mock_data_array.h @@ -2,6 +2,8 @@ #include "../data_array.h" +namespace extractor { + class MockDataArray : public DataArray { public: MOCK_CONST_METHOD0(GetData, const vector&()); @@ -17,3 +19,5 @@ class MockDataArray : public DataArray { MOCK_CONST_METHOD1(GetSentenceStart, int(int sentence_id)); MOCK_CONST_METHOD1(GetSentenceId, int(int position)); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_fast_intersector.h b/extractor/mocks/mock_fast_intersector.h index 201386f2..665add65 100644 --- a/extractor/mocks/mock_fast_intersector.h +++ b/extractor/mocks/mock_fast_intersector.h @@ -4,8 +4,12 @@ #include "../phrase.h" #include "../phrase_location.h" +namespace extractor { + class MockFastIntersector : public FastIntersector { public: MOCK_METHOD3(Intersect, PhraseLocation(PhraseLocation&, PhraseLocation&, const Phrase&)); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_feature.h b/extractor/mocks/mock_feature.h index d2137629..19ba4de9 100644 --- a/extractor/mocks/mock_feature.h +++ b/extractor/mocks/mock_feature.h @@ -2,8 +2,14 @@ #include "../features/feature.h" +namespace extractor { +namespace features { + class MockFeature : public Feature { public: MOCK_CONST_METHOD1(Score, double(const FeatureContext& context)); MOCK_CONST_METHOD0(GetName, string()); }; + +} // namespace features +} // namespace extractor diff --git a/extractor/mocks/mock_matchings_finder.h b/extractor/mocks/mock_matchings_finder.h index 3e80d266..ffbb06c7 100644 --- a/extractor/mocks/mock_matchings_finder.h +++ b/extractor/mocks/mock_matchings_finder.h @@ -3,7 +3,11 @@ #include "../matchings_finder.h" #include "../phrase_location.h" +namespace extractor { + class MockMatchingsFinder : public MatchingsFinder { public: MOCK_METHOD3(Find, PhraseLocation(PhraseLocation&, const string&, int)); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_precomputation.h b/extractor/mocks/mock_precomputation.h index 9bc72235..64934b94 100644 --- a/extractor/mocks/mock_precomputation.h +++ b/extractor/mocks/mock_precomputation.h @@ -2,7 +2,11 @@ #include "../precomputation.h" +namespace extractor { + class MockPrecomputation : public Precomputation { public: MOCK_CONST_METHOD0(GetCollocations, const Index&()); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_rule_extractor.h b/extractor/mocks/mock_rule_extractor.h index f18e009a..28b644b0 100644 --- a/extractor/mocks/mock_rule_extractor.h +++ b/extractor/mocks/mock_rule_extractor.h @@ -5,8 +5,12 @@ #include "../rule.h" #include "../rule_extractor.h" +namespace extractor { + class MockRuleExtractor : public RuleExtractor { public: MOCK_CONST_METHOD2(ExtractRules, vector(const Phrase&, const PhraseLocation&)); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_rule_extractor_helper.h b/extractor/mocks/mock_rule_extractor_helper.h index 63ff1048..3b0ac0f5 100644 --- a/extractor/mocks/mock_rule_extractor_helper.h +++ b/extractor/mocks/mock_rule_extractor_helper.h @@ -6,6 +6,8 @@ using namespace std; +namespace extractor { + typedef unordered_map Indexes; class MockRuleExtractorHelper : public RuleExtractorHelper { @@ -76,3 +78,5 @@ class MockRuleExtractorHelper : public RuleExtractorHelper { bool met_constraints; bool get_gaps; }; + +} // namespace extractor diff --git a/extractor/mocks/mock_rule_factory.h b/extractor/mocks/mock_rule_factory.h index 2a96be93..11cb9ab5 100644 --- a/extractor/mocks/mock_rule_factory.h +++ b/extractor/mocks/mock_rule_factory.h @@ -3,7 +3,11 @@ #include "../grammar.h" #include "../rule_factory.h" +namespace extractor { + class MockHieroCachingRuleFactory : public HieroCachingRuleFactory { public: MOCK_METHOD1(GetGrammar, Grammar(const vector& word_ids)); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_sampler.h b/extractor/mocks/mock_sampler.h index b2306109..7022f7b3 100644 --- a/extractor/mocks/mock_sampler.h +++ b/extractor/mocks/mock_sampler.h @@ -3,7 +3,11 @@ #include "../phrase_location.h" #include "../sampler.h" +namespace extractor { + class MockSampler : public Sampler { public: MOCK_CONST_METHOD1(Sample, PhraseLocation(const PhraseLocation& location)); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_scorer.h b/extractor/mocks/mock_scorer.h index 48115ef4..4d593ddf 100644 --- a/extractor/mocks/mock_scorer.h +++ b/extractor/mocks/mock_scorer.h @@ -3,8 +3,13 @@ #include "../scorer.h" #include "../features/feature.h" +namespace extractor { + class MockScorer : public Scorer { public: - MOCK_CONST_METHOD1(Score, vector(const FeatureContext& context)); + MOCK_CONST_METHOD1(Score, vector( + const features::FeatureContext& context)); MOCK_CONST_METHOD0(GetFeatureNames, vector()); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_suffix_array.h b/extractor/mocks/mock_suffix_array.h index 11a3a443..6886232a 100644 --- a/extractor/mocks/mock_suffix_array.h +++ b/extractor/mocks/mock_suffix_array.h @@ -9,6 +9,8 @@ using namespace std; +namespace extractor { + class MockSuffixArray : public SuffixArray { public: MOCK_CONST_METHOD0(GetSize, int()); @@ -17,3 +19,5 @@ class MockSuffixArray : public SuffixArray { MOCK_CONST_METHOD1(GetSuffix, int(int)); MOCK_CONST_METHOD4(Lookup, PhraseLocation(int, int, const string& word, int)); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_target_phrase_extractor.h b/extractor/mocks/mock_target_phrase_extractor.h index 6dc6bba6..e5e9aeab 100644 --- a/extractor/mocks/mock_target_phrase_extractor.h +++ b/extractor/mocks/mock_target_phrase_extractor.h @@ -2,6 +2,8 @@ #include "../target_phrase_extractor.h" +namespace extractor { + typedef pair PhraseExtract; class MockTargetPhraseExtractor : public TargetPhraseExtractor { @@ -10,3 +12,5 @@ class MockTargetPhraseExtractor : public TargetPhraseExtractor { const vector > &, const vector&, int, int, const unordered_map&, int)); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_translation_table.h b/extractor/mocks/mock_translation_table.h index a35c9327..358c854f 100644 --- a/extractor/mocks/mock_translation_table.h +++ b/extractor/mocks/mock_translation_table.h @@ -2,8 +2,12 @@ #include "../translation_table.h" +namespace extractor { + class MockTranslationTable : public TranslationTable { public: MOCK_METHOD2(GetSourceGivenTargetScore, double(const string&, const string&)); MOCK_METHOD2(GetTargetGivenSourceScore, double(const string&, const string&)); }; + +} // namespace extractor diff --git a/extractor/mocks/mock_vocabulary.h b/extractor/mocks/mock_vocabulary.h index e5c191f5..802c29b4 100644 --- a/extractor/mocks/mock_vocabulary.h +++ b/extractor/mocks/mock_vocabulary.h @@ -2,8 +2,12 @@ #include "../vocabulary.h" +namespace extractor { + class MockVocabulary : public Vocabulary { public: MOCK_METHOD1(GetTerminalValue, string(int word_id)); MOCK_METHOD1(GetTerminalIndex, int(const string& word)); }; + +} // namespace extractor diff --git a/extractor/phrase.cc b/extractor/phrase.cc index 6dc242db..244fab07 100644 --- a/extractor/phrase.cc +++ b/extractor/phrase.cc @@ -1,5 +1,7 @@ #include "phrase.h" +namespace extractor { + int Phrase::Arity() const { return var_pos.size(); } @@ -52,3 +54,5 @@ ostream& operator<<(ostream& os, const Phrase& phrase) { } return os; } + +} // namspace extractor diff --git a/extractor/phrase.h b/extractor/phrase.h index f40a8169..8c98a025 100644 --- a/extractor/phrase.h +++ b/extractor/phrase.h @@ -9,6 +9,8 @@ using namespace std; +namespace extractor { + class Phrase { public: friend Phrase PhraseBuilder::Build(const vector& phrase); @@ -38,4 +40,6 @@ class Phrase { vector words; }; +} // namespace extractor + #endif diff --git a/extractor/phrase_builder.cc b/extractor/phrase_builder.cc index 4325390c..9faee4be 100644 --- a/extractor/phrase_builder.cc +++ b/extractor/phrase_builder.cc @@ -3,6 +3,8 @@ #include "phrase.h" #include "vocabulary.h" +namespace extractor { + PhraseBuilder::PhraseBuilder(shared_ptr vocabulary) : vocabulary(vocabulary) {} @@ -42,3 +44,5 @@ Phrase PhraseBuilder::Extend(const Phrase& phrase, bool start_x, bool end_x) { return Build(symbols); } + +} // namespace extractor diff --git a/extractor/phrase_builder.h b/extractor/phrase_builder.h index a49af457..2956fd35 100644 --- a/extractor/phrase_builder.h +++ b/extractor/phrase_builder.h @@ -6,6 +6,8 @@ using namespace std; +namespace extractor { + class Phrase; class Vocabulary; @@ -21,4 +23,6 @@ class PhraseBuilder { shared_ptr vocabulary; }; +} // namespace extractor + #endif diff --git a/extractor/phrase_location.cc b/extractor/phrase_location.cc index b0bfed80..678ae270 100644 --- a/extractor/phrase_location.cc +++ b/extractor/phrase_location.cc @@ -1,5 +1,7 @@ #include "phrase_location.h" +namespace extractor { + PhraseLocation::PhraseLocation(int sa_low, int sa_high) : sa_low(sa_low), sa_high(sa_high), num_subpatterns(0) {} @@ -37,3 +39,5 @@ bool operator==(const PhraseLocation& a, const PhraseLocation& b) { return *a.matchings == *b.matchings; } + +} // namespace extractor diff --git a/extractor/phrase_location.h b/extractor/phrase_location.h index a0eb36c8..e5f3cf08 100644 --- a/extractor/phrase_location.h +++ b/extractor/phrase_location.h @@ -6,6 +6,8 @@ using namespace std; +namespace extractor { + struct PhraseLocation { PhraseLocation(int sa_low = -1, int sa_high = -1); @@ -22,4 +24,6 @@ struct PhraseLocation { int num_subpatterns; }; +} // namespace extractor + #endif diff --git a/extractor/phrase_test.cc b/extractor/phrase_test.cc index 2b553b6f..c8176178 100644 --- a/extractor/phrase_test.cc +++ b/extractor/phrase_test.cc @@ -10,6 +10,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class PhraseTest : public Test { @@ -58,4 +59,5 @@ TEST_F(PhraseTest, TestGetSymbol) { } } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/precomputation.cc b/extractor/precomputation.cc index 189ac42c..8cc32ffd 100644 --- a/extractor/precomputation.cc +++ b/extractor/precomputation.cc @@ -8,6 +8,8 @@ using namespace std; +namespace extractor { + int Precomputation::NON_TERMINAL = -1; Precomputation::Precomputation( @@ -173,3 +175,5 @@ void Precomputation::WriteBinary(const fs::path& filepath) const { const Index& Precomputation::GetCollocations() const { return collocations; } + +} // namespace extractor diff --git a/extractor/precomputation.h b/extractor/precomputation.h index 3d44c2a6..dbd99c14 100644 --- a/extractor/precomputation.h +++ b/extractor/precomputation.h @@ -13,11 +13,13 @@ namespace fs = boost::filesystem; using namespace std; -class SuffixArray; +namespace extractor { typedef boost::hash > VectorHash; typedef unordered_map, vector, VectorHash> Index; +class SuffixArray; + class Precomputation { public: Precomputation( @@ -51,4 +53,6 @@ class Precomputation { Index collocations; }; +} // namespace extractor + #endif diff --git a/extractor/precomputation_test.cc b/extractor/precomputation_test.cc index 6b77b9c0..04e3850d 100644 --- a/extractor/precomputation_test.cc +++ b/extractor/precomputation_test.cc @@ -10,6 +10,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class PrecomputationTest : public Test { @@ -101,4 +102,6 @@ TEST_F(PrecomputationTest, TestCollocations) { EXPECT_EQ(0, collocations.count(key)); } -} // namespace +} // namespace +} // namespace extractor + diff --git a/extractor/rule.cc b/extractor/rule.cc index 9c7ac9b5..b6c7d783 100644 --- a/extractor/rule.cc +++ b/extractor/rule.cc @@ -1,5 +1,7 @@ #include "rule.h" +namespace extractor { + Rule::Rule(const Phrase& source_phrase, const Phrase& target_phrase, const vector& scores, @@ -8,3 +10,5 @@ Rule::Rule(const Phrase& source_phrase, target_phrase(target_phrase), scores(scores), alignment(alignment) {} + +} // namespace extractor diff --git a/extractor/rule.h b/extractor/rule.h index 64ff8794..b4d45fc1 100644 --- a/extractor/rule.h +++ b/extractor/rule.h @@ -7,6 +7,8 @@ using namespace std; +namespace extractor { + struct Rule { Rule(const Phrase& source_phrase, const Phrase& target_phrase, const vector& scores, const vector >& alignment); @@ -17,4 +19,6 @@ struct Rule { vector > alignment; }; +} // namespace extractor + #endif diff --git a/extractor/rule_extractor.cc b/extractor/rule_extractor.cc index 92343241..b9286472 100644 --- a/extractor/rule_extractor.cc +++ b/extractor/rule_extractor.cc @@ -14,6 +14,8 @@ using namespace std; +namespace extractor { + RuleExtractor::RuleExtractor( shared_ptr source_data_array, shared_ptr target_data_array, @@ -106,7 +108,7 @@ vector RuleExtractor::ExtractRules(const Phrase& phrase, } } - FeatureContext context(source_phrase, target_phrase, + features::FeatureContext context(source_phrase, target_phrase, source_phrase_counter[source_phrase], num_locations, num_samples); vector scores = scorer->Score(context); rules.push_back(Rule(source_phrase, target_phrase, scores, @@ -313,3 +315,5 @@ void RuleExtractor::AddNonterminalExtremities( AddExtracts(extracts, new_source_phrase, source_indexes, target_gaps, target_low, target_x_low, target_x_high, sentence_id); } + +} // namespace extractor diff --git a/extractor/rule_extractor.h b/extractor/rule_extractor.h index a087dc6d..8b6daeea 100644 --- a/extractor/rule_extractor.h +++ b/extractor/rule_extractor.h @@ -9,6 +9,10 @@ using namespace std; +namespace extractor { + +typedef vector > PhraseAlignment; + class Alignment; class DataArray; class PhraseBuilder; @@ -18,8 +22,6 @@ class RuleExtractorHelper; class Scorer; class TargetPhraseExtractor; -typedef vector > PhraseAlignment; - struct Extract { Extract(const Phrase& source_phrase, const Phrase& target_phrase, double pairs_count, const PhraseAlignment& alignment) : @@ -101,4 +103,6 @@ class RuleExtractor { bool require_tight_phrases; }; +} // namespace extractor + #endif diff --git a/extractor/rule_extractor_helper.cc b/extractor/rule_extractor_helper.cc index ed6ae3a1..553b56d4 100644 --- a/extractor/rule_extractor_helper.cc +++ b/extractor/rule_extractor_helper.cc @@ -3,6 +3,8 @@ #include "data_array.h" #include "alignment.h" +namespace extractor { + RuleExtractorHelper::RuleExtractorHelper( shared_ptr source_data_array, shared_ptr target_data_array, @@ -354,3 +356,5 @@ unordered_map RuleExtractorHelper::GetSourceIndexes( } return source_indexes; } + +} // namespace extractor diff --git a/extractor/rule_extractor_helper.h b/extractor/rule_extractor_helper.h index 3478bfc8..95274df6 100644 --- a/extractor/rule_extractor_helper.h +++ b/extractor/rule_extractor_helper.h @@ -7,6 +7,8 @@ using namespace std; +namespace extractor { + class Alignment; class DataArray; @@ -79,4 +81,6 @@ class RuleExtractorHelper { bool require_tight_phrases; }; +} // namespace extractor + #endif diff --git a/extractor/rule_extractor_helper_test.cc b/extractor/rule_extractor_helper_test.cc index 29213312..ec0635b1 100644 --- a/extractor/rule_extractor_helper_test.cc +++ b/extractor/rule_extractor_helper_test.cc @@ -9,6 +9,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class RuleExtractorHelperTest : public Test { @@ -619,4 +620,5 @@ TEST_F(RuleExtractorHelperTest, TestGetGapIntegrityChecksFailed) { met_constraints)); } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/rule_extractor_test.cc b/extractor/rule_extractor_test.cc index 0be44d4d..1b543fc9 100644 --- a/extractor/rule_extractor_test.cc +++ b/extractor/rule_extractor_test.cc @@ -17,6 +17,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class RuleExtractorTest : public Test { @@ -163,4 +164,5 @@ TEST_F(RuleExtractorTest, TestExtractRulesAddExtremities) { EXPECT_EQ(4, rules.size()); } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/rule_factory.cc b/extractor/rule_factory.cc index 4908dac0..51f85c30 100644 --- a/extractor/rule_factory.cc +++ b/extractor/rule_factory.cc @@ -9,6 +9,7 @@ #include "fast_intersector.h" #include "matchings_finder.h" #include "phrase.h" +#include "phrase_builder.h" #include "rule.h" #include "rule_extractor.h" #include "sampler.h" @@ -20,6 +21,8 @@ using namespace std; using namespace chrono; +namespace extractor { + typedef high_resolution_clock Clock; struct State { @@ -282,3 +285,5 @@ vector HieroCachingRuleFactory::ExtendState( return new_states; } + +} // namespace extractor diff --git a/extractor/rule_factory.h b/extractor/rule_factory.h index 2d1c7f5a..0de04e40 100644 --- a/extractor/rule_factory.h +++ b/extractor/rule_factory.h @@ -5,15 +5,17 @@ #include #include "matchings_trie.h" -#include "phrase_builder.h" using namespace std; +namespace extractor { + class Alignment; class DataArray; +class FastIntersector; class Grammar; class MatchingsFinder; -class FastIntersector; +class PhraseBuilder; class Precomputation; class Rule; class RuleExtractor; @@ -92,4 +94,6 @@ class HieroCachingRuleFactory { int max_rule_symbols; }; +} // namespace extractor + #endif diff --git a/extractor/rule_factory_test.cc b/extractor/rule_factory_test.cc index fc709461..2129dfa0 100644 --- a/extractor/rule_factory_test.cc +++ b/extractor/rule_factory_test.cc @@ -18,6 +18,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class RuleFactoryTest : public Test { @@ -98,4 +99,5 @@ TEST_F(RuleFactoryTest, TestGetGrammarRepeatingWords) { EXPECT_EQ(28, grammar.GetRules().size()); } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/run_extractor.cc b/extractor/run_extractor.cc index 5255737d..c701c8d0 100644 --- a/extractor/run_extractor.cc +++ b/extractor/run_extractor.cc @@ -1,6 +1,7 @@ #include #include #include +#include #include #include @@ -30,6 +31,8 @@ namespace fs = boost::filesystem; namespace po = boost::program_options; using namespace std; +using namespace extractor; +using namespace features; int main(int argc, char** argv) { // TODO(pauldb): Also take arguments from config file. @@ -146,13 +149,13 @@ int main(int argc, char** argv) { Clock::time_point extraction_start_time = Clock::now(); vector > features = { - make_shared(), - make_shared(), - make_shared(), - make_shared(table), - make_shared(table), - make_shared(), - make_shared() +// make_shared(), +// make_shared(), +// make_shared(), +// make_shared(table), +// make_shared(table), +// make_shared(), +// make_shared() }; shared_ptr scorer = make_shared(features); diff --git a/extractor/sampler.cc b/extractor/sampler.cc index 5067ca8a..d128913f 100644 --- a/extractor/sampler.cc +++ b/extractor/sampler.cc @@ -3,6 +3,8 @@ #include "phrase_location.h" #include "suffix_array.h" +namespace extractor { + Sampler::Sampler(shared_ptr suffix_array, int max_samples) : suffix_array(suffix_array), max_samples(max_samples) {} @@ -39,3 +41,5 @@ int Sampler::Round(double x) const { // TODO(pauldb): Remove EPS. return x + 0.5 + 1e-8; } + +} // namespace extractor diff --git a/extractor/sampler.h b/extractor/sampler.h index 9cf321fb..cda28b10 100644 --- a/extractor/sampler.h +++ b/extractor/sampler.h @@ -5,6 +5,8 @@ using namespace std; +namespace extractor { + class PhraseLocation; class SuffixArray; @@ -26,4 +28,6 @@ class Sampler { int max_samples; }; +} // namespace extractor + #endif diff --git a/extractor/sampler_test.cc b/extractor/sampler_test.cc index 4f91965b..e9abebfa 100644 --- a/extractor/sampler_test.cc +++ b/extractor/sampler_test.cc @@ -9,6 +9,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class SamplerTest : public Test { @@ -69,4 +70,5 @@ TEST_F(SamplerTest, TestSubstringsSample) { EXPECT_EQ(PhraseLocation(expected_locations, 2), sampler->Sample(location)); } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/scorer.cc b/extractor/scorer.cc index f28b3181..d3ebf1c9 100644 --- a/extractor/scorer.cc +++ b/extractor/scorer.cc @@ -2,14 +2,16 @@ #include "features/feature.h" -Scorer::Scorer(const vector >& features) : +namespace extractor { + +Scorer::Scorer(const vector >& features) : features(features) {} Scorer::Scorer() {} Scorer::~Scorer() {} -vector Scorer::Score(const FeatureContext& context) const { +vector Scorer::Score(const features::FeatureContext& context) const { vector scores; for (auto feature: features) { scores.push_back(feature->Score(context)); @@ -24,3 +26,5 @@ vector Scorer::GetFeatureNames() const { } return feature_names; } + +} // namespace extractor diff --git a/extractor/scorer.h b/extractor/scorer.h index ba71a6ee..c31db0ca 100644 --- a/extractor/scorer.h +++ b/extractor/scorer.h @@ -7,16 +7,20 @@ using namespace std; -class Feature; -class FeatureContext; +namespace extractor { + +namespace features { + class Feature; + class FeatureContext; +} // namespace features class Scorer { public: - Scorer(const vector >& features); + Scorer(const vector >& features); virtual ~Scorer(); - virtual vector Score(const FeatureContext& context) const; + virtual vector Score(const features::FeatureContext& context) const; virtual vector GetFeatureNames() const; @@ -24,7 +28,9 @@ class Scorer { Scorer(); private: - vector > features; + vector > features; }; +} // namespace extractor + #endif diff --git a/extractor/scorer_test.cc b/extractor/scorer_test.cc index 56a85762..3a09c9cc 100644 --- a/extractor/scorer_test.cc +++ b/extractor/scorer_test.cc @@ -10,32 +10,33 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class ScorerTest : public Test { protected: virtual void SetUp() { - feature1 = make_shared(); + feature1 = make_shared(); EXPECT_CALL(*feature1, Score(_)).WillRepeatedly(Return(0.5)); EXPECT_CALL(*feature1, GetName()).WillRepeatedly(Return("f1")); - feature2 = make_shared(); + feature2 = make_shared(); EXPECT_CALL(*feature2, Score(_)).WillRepeatedly(Return(-1.3)); EXPECT_CALL(*feature2, GetName()).WillRepeatedly(Return("f2")); - vector > features = {feature1, feature2}; + vector > features = {feature1, feature2}; scorer = make_shared(features); } - shared_ptr feature1; - shared_ptr feature2; + shared_ptr feature1; + shared_ptr feature2; shared_ptr scorer; }; TEST_F(ScorerTest, TestScore) { vector expected_scores = {0.5, -1.3}; Phrase phrase; - FeatureContext context(phrase, phrase, 0.3, 2, 11); + features::FeatureContext context(phrase, phrase, 0.3, 2, 11); EXPECT_EQ(expected_scores, scorer->Score(context)); } @@ -44,4 +45,5 @@ TEST_F(ScorerTest, TestGetNames) { EXPECT_EQ(expected_names, scorer->GetFeatureNames()); } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/suffix_array.cc b/extractor/suffix_array.cc index ab8a0913..9988b1a2 100644 --- a/extractor/suffix_array.cc +++ b/extractor/suffix_array.cc @@ -13,6 +13,8 @@ namespace fs = boost::filesystem; using namespace std; using namespace chrono; +namespace extractor { + SuffixArray::SuffixArray(shared_ptr data_array) : data_array(data_array) { BuildSuffixArray(); @@ -227,3 +229,5 @@ int SuffixArray::LookupRangeStart(int low, int high, int word_id, } return result; } + +} // namespace extractor diff --git a/extractor/suffix_array.h b/extractor/suffix_array.h index 79a22694..7a4f1110 100644 --- a/extractor/suffix_array.h +++ b/extractor/suffix_array.h @@ -10,6 +10,8 @@ namespace fs = boost::filesystem; using namespace std; +namespace extractor { + class DataArray; class PhraseLocation; @@ -51,4 +53,6 @@ class SuffixArray { vector word_start; }; +} // namespace extractor + #endif diff --git a/extractor/suffix_array_test.cc b/extractor/suffix_array_test.cc index 60295567..8431a16e 100644 --- a/extractor/suffix_array_test.cc +++ b/extractor/suffix_array_test.cc @@ -9,6 +9,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class SuffixArrayTest : public Test { @@ -73,4 +74,5 @@ TEST_F(SuffixArrayTest, TestLookup) { EXPECT_EQ(PhraseLocation(11, 11), suffix_array->Lookup(11, 13, "word5", 1)); } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/target_phrase_extractor.cc b/extractor/target_phrase_extractor.cc index ac583953..9f8bc6e2 100644 --- a/extractor/target_phrase_extractor.cc +++ b/extractor/target_phrase_extractor.cc @@ -11,6 +11,8 @@ using namespace std; +namespace extractor { + TargetPhraseExtractor::TargetPhraseExtractor( shared_ptr target_data_array, shared_ptr alignment, @@ -142,3 +144,5 @@ void TargetPhraseExtractor::GeneratePhrases( ++subpatterns[index]; } } + +} // namespace extractor diff --git a/extractor/target_phrase_extractor.h b/extractor/target_phrase_extractor.h index 134f24cc..a4b54145 100644 --- a/extractor/target_phrase_extractor.h +++ b/extractor/target_phrase_extractor.h @@ -7,6 +7,10 @@ using namespace std; +namespace extractor { + +typedef vector > PhraseAlignment; + class Alignment; class DataArray; class Phrase; @@ -14,8 +18,6 @@ class PhraseBuilder; class RuleExtractorHelper; class Vocabulary; -typedef vector > PhraseAlignment; - class TargetPhraseExtractor { public: TargetPhraseExtractor(shared_ptr target_data_array, @@ -53,4 +55,6 @@ class TargetPhraseExtractor { bool require_tight_phrases; }; +} // namespace extractor + #endif diff --git a/extractor/target_phrase_extractor_test.cc b/extractor/target_phrase_extractor_test.cc index 7394f4d9..51c753ee 100644 --- a/extractor/target_phrase_extractor_test.cc +++ b/extractor/target_phrase_extractor_test.cc @@ -14,6 +14,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { class TargetPhraseExtractorTest : public Test { @@ -113,4 +114,5 @@ TEST_F(TargetPhraseExtractorTest, TestExtractPhrasesTightPhrasesFalse) { // look like. } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/time_util.cc b/extractor/time_util.cc index 88395f77..e46a0c3d 100644 --- a/extractor/time_util.cc +++ b/extractor/time_util.cc @@ -1,6 +1,10 @@ #include "time_util.h" +namespace extractor { + double GetDuration(const Clock::time_point& start_time, const Clock::time_point& stop_time) { return duration_cast(stop_time - start_time).count() / 1000.0; } + +} // namespace extractor diff --git a/extractor/time_util.h b/extractor/time_util.h index 6f7eda70..45f79199 100644 --- a/extractor/time_util.h +++ b/extractor/time_util.h @@ -6,9 +6,13 @@ using namespace std; using namespace chrono; +namespace extractor { + typedef high_resolution_clock Clock; double GetDuration(const Clock::time_point& start_time, const Clock::time_point& stop_time); +} // namespace extractor + #endif diff --git a/extractor/translation_table.cc b/extractor/translation_table.cc index a48c0657..1852a357 100644 --- a/extractor/translation_table.cc +++ b/extractor/translation_table.cc @@ -10,6 +10,8 @@ using namespace std; +namespace extractor { + TranslationTable::TranslationTable(shared_ptr source_data_array, shared_ptr target_data_array, shared_ptr alignment) : @@ -115,3 +117,5 @@ void TranslationTable::WriteBinary(const fs::path& filepath) const { fwrite(&entry.second, sizeof(entry.second), 1, file); } } + +} // namespace extractor diff --git a/extractor/translation_table.h b/extractor/translation_table.h index 157ad3af..a7be26f5 100644 --- a/extractor/translation_table.h +++ b/extractor/translation_table.h @@ -11,11 +11,13 @@ using namespace std; namespace fs = boost::filesystem; -class Alignment; -class DataArray; +namespace extractor { typedef boost::hash > PairHash; +class Alignment; +class DataArray; + class TranslationTable { public: TranslationTable( @@ -50,4 +52,6 @@ class TranslationTable { translation_probabilities; }; +} // namespace extractor + #endif diff --git a/extractor/translation_table_test.cc b/extractor/translation_table_test.cc index c99f3f93..051b5715 100644 --- a/extractor/translation_table_test.cc +++ b/extractor/translation_table_test.cc @@ -11,6 +11,7 @@ using namespace std; using namespace ::testing; +namespace extractor { namespace { TEST(TranslationTableTest, TestScores) { @@ -79,4 +80,5 @@ TEST(TranslationTableTest, TestScores) { EXPECT_EQ(-1, table->GetSourceGivenTargetScore("c", "d")); } -} // namespace +} // namespace +} // namespace extractor diff --git a/extractor/vocabulary.cc b/extractor/vocabulary.cc index b68d76a9..57f564d9 100644 --- a/extractor/vocabulary.cc +++ b/extractor/vocabulary.cc @@ -1,5 +1,7 @@ #include "vocabulary.h" +namespace extractor { + Vocabulary::~Vocabulary() {} int Vocabulary::GetTerminalIndex(const string& word) { @@ -29,3 +31,4 @@ int Vocabulary::Size() { return words.size(); } +} // namespace extractor diff --git a/extractor/vocabulary.h b/extractor/vocabulary.h index ff3e7a63..dcc2a8fa 100644 --- a/extractor/vocabulary.h +++ b/extractor/vocabulary.h @@ -7,6 +7,8 @@ using namespace std; +namespace extractor { + class Vocabulary { public: virtual ~Vocabulary(); @@ -26,4 +28,6 @@ class Vocabulary { vector words; }; +} // namespace extractor + #endif -- cgit v1.2.3 From 41b2aae5cf08899da5c46e944ead4ddf6639a3f0 Mon Sep 17 00:00:00 2001 From: Paul Baltescu Date: Wed, 6 Mar 2013 17:13:13 +0000 Subject: Fix broken test & includes. --- extractor/features/feature.h | 3 +-- extractor/features/max_lex_source_given_target.cc | 4 ++-- .../features/max_lex_source_given_target_test.cc | 8 ++++---- extractor/features/max_lex_target_given_source.cc | 4 ++-- .../features/max_lex_target_given_source_test.cc | 8 ++++---- extractor/mocks/mock_alignment.h | 2 +- extractor/mocks/mock_data_array.h | 2 +- extractor/mocks/mock_fast_intersector.h | 6 +++--- extractor/mocks/mock_feature.h | 2 +- extractor/mocks/mock_matchings_finder.h | 4 ++-- extractor/mocks/mock_precomputation.h | 2 +- extractor/mocks/mock_rule_extractor.h | 8 ++++---- extractor/mocks/mock_rule_extractor_helper.h | 2 +- extractor/mocks/mock_rule_factory.h | 4 ++-- extractor/mocks/mock_sampler.h | 4 ++-- extractor/mocks/mock_scorer.h | 4 ++-- extractor/mocks/mock_suffix_array.h | 6 +++--- extractor/mocks/mock_target_phrase_extractor.h | 2 +- extractor/mocks/mock_translation_table.h | 2 +- extractor/mocks/mock_vocabulary.h | 2 +- extractor/precomputation_test.cc | 19 +++++++++---------- 21 files changed, 48 insertions(+), 50 deletions(-) (limited to 'extractor/features/max_lex_target_given_source.cc') diff --git a/extractor/features/feature.h b/extractor/features/feature.h index de2827bc..6693ccbf 100644 --- a/extractor/features/feature.h +++ b/extractor/features/feature.h @@ -3,8 +3,7 @@ #include -//TODO(pauldb): include headers nicely. -#include "../phrase.h" +#include "phrase.h" using namespace std; diff --git a/extractor/features/max_lex_source_given_target.cc b/extractor/features/max_lex_source_given_target.cc index 3ffe598c..65d0ec68 100644 --- a/extractor/features/max_lex_source_given_target.cc +++ b/extractor/features/max_lex_source_given_target.cc @@ -2,8 +2,8 @@ #include -#include "../data_array.h" -#include "../translation_table.h" +#include "data_array.h" +#include "translation_table.h" namespace extractor { namespace features { diff --git a/extractor/features/max_lex_source_given_target_test.cc b/extractor/features/max_lex_source_given_target_test.cc index c1edb483..7f6aae41 100644 --- a/extractor/features/max_lex_source_given_target_test.cc +++ b/extractor/features/max_lex_source_given_target_test.cc @@ -4,10 +4,10 @@ #include #include -#include "../mocks/mock_translation_table.h" -#include "../mocks/mock_vocabulary.h" -#include "../data_array.h" -#include "../phrase_builder.h" +#include "data_array.h" +#include "mocks/mock_translation_table.h" +#include "mocks/mock_vocabulary.h" +#include "phrase_builder.h" #include "max_lex_source_given_target.h" using namespace std; diff --git a/extractor/features/max_lex_target_given_source.cc b/extractor/features/max_lex_target_given_source.cc index 30140d80..33783054 100644 --- a/extractor/features/max_lex_target_given_source.cc +++ b/extractor/features/max_lex_target_given_source.cc @@ -2,8 +2,8 @@ #include -#include "../data_array.h" -#include "../translation_table.h" +#include "data_array.h" +#include "translation_table.h" namespace extractor { namespace features { diff --git a/extractor/features/max_lex_target_given_source_test.cc b/extractor/features/max_lex_target_given_source_test.cc index 9ceb13e5..6d0efd9c 100644 --- a/extractor/features/max_lex_target_given_source_test.cc +++ b/extractor/features/max_lex_target_given_source_test.cc @@ -4,10 +4,10 @@ #include #include -#include "../mocks/mock_translation_table.h" -#include "../mocks/mock_vocabulary.h" -#include "../data_array.h" -#include "../phrase_builder.h" +#include "data_array.h" +#include "mocks/mock_translation_table.h" +#include "mocks/mock_vocabulary.h" +#include "phrase_builder.h" #include "max_lex_target_given_source.h" using namespace std; diff --git a/extractor/mocks/mock_alignment.h b/extractor/mocks/mock_alignment.h index 3d745e9d..299c3d1c 100644 --- a/extractor/mocks/mock_alignment.h +++ b/extractor/mocks/mock_alignment.h @@ -1,6 +1,6 @@ #include -#include "../alignment.h" +#include "alignment.h" namespace extractor { diff --git a/extractor/mocks/mock_data_array.h b/extractor/mocks/mock_data_array.h index cf9f3671..6f85abb4 100644 --- a/extractor/mocks/mock_data_array.h +++ b/extractor/mocks/mock_data_array.h @@ -1,6 +1,6 @@ #include -#include "../data_array.h" +#include "data_array.h" namespace extractor { diff --git a/extractor/mocks/mock_fast_intersector.h b/extractor/mocks/mock_fast_intersector.h index 665add65..f0b628d7 100644 --- a/extractor/mocks/mock_fast_intersector.h +++ b/extractor/mocks/mock_fast_intersector.h @@ -1,8 +1,8 @@ #include -#include "../fast_intersector.h" -#include "../phrase.h" -#include "../phrase_location.h" +#include "fast_intersector.h" +#include "phrase.h" +#include "phrase_location.h" namespace extractor { diff --git a/extractor/mocks/mock_feature.h b/extractor/mocks/mock_feature.h index 19ba4de9..0b0f0ead 100644 --- a/extractor/mocks/mock_feature.h +++ b/extractor/mocks/mock_feature.h @@ -1,6 +1,6 @@ #include -#include "../features/feature.h" +#include "features/feature.h" namespace extractor { namespace features { diff --git a/extractor/mocks/mock_matchings_finder.h b/extractor/mocks/mock_matchings_finder.h index ffbb06c7..827526fd 100644 --- a/extractor/mocks/mock_matchings_finder.h +++ b/extractor/mocks/mock_matchings_finder.h @@ -1,7 +1,7 @@ #include -#include "../matchings_finder.h" -#include "../phrase_location.h" +#include "matchings_finder.h" +#include "phrase_location.h" namespace extractor { diff --git a/extractor/mocks/mock_precomputation.h b/extractor/mocks/mock_precomputation.h index 64934b94..8753343e 100644 --- a/extractor/mocks/mock_precomputation.h +++ b/extractor/mocks/mock_precomputation.h @@ -1,6 +1,6 @@ #include -#include "../precomputation.h" +#include "precomputation.h" namespace extractor { diff --git a/extractor/mocks/mock_rule_extractor.h b/extractor/mocks/mock_rule_extractor.h index 28b644b0..aad11651 100644 --- a/extractor/mocks/mock_rule_extractor.h +++ b/extractor/mocks/mock_rule_extractor.h @@ -1,9 +1,9 @@ #include -#include "../phrase.h" -#include "../phrase_builder.h" -#include "../rule.h" -#include "../rule_extractor.h" +#include "phrase.h" +#include "phrase_builder.h" +#include "rule.h" +#include "rule_extractor.h" namespace extractor { diff --git a/extractor/mocks/mock_rule_extractor_helper.h b/extractor/mocks/mock_rule_extractor_helper.h index 3b0ac0f5..cf196ab5 100644 --- a/extractor/mocks/mock_rule_extractor_helper.h +++ b/extractor/mocks/mock_rule_extractor_helper.h @@ -2,7 +2,7 @@ #include -#include "../rule_extractor_helper.h" +#include "rule_extractor_helper.h" using namespace std; diff --git a/extractor/mocks/mock_rule_factory.h b/extractor/mocks/mock_rule_factory.h index 11cb9ab5..7389b396 100644 --- a/extractor/mocks/mock_rule_factory.h +++ b/extractor/mocks/mock_rule_factory.h @@ -1,7 +1,7 @@ #include -#include "../grammar.h" -#include "../rule_factory.h" +#include "grammar.h" +#include "rule_factory.h" namespace extractor { diff --git a/extractor/mocks/mock_sampler.h b/extractor/mocks/mock_sampler.h index 7022f7b3..75c43c27 100644 --- a/extractor/mocks/mock_sampler.h +++ b/extractor/mocks/mock_sampler.h @@ -1,7 +1,7 @@ #include -#include "../phrase_location.h" -#include "../sampler.h" +#include "phrase_location.h" +#include "sampler.h" namespace extractor { diff --git a/extractor/mocks/mock_scorer.h b/extractor/mocks/mock_scorer.h index 4d593ddf..cc0c444d 100644 --- a/extractor/mocks/mock_scorer.h +++ b/extractor/mocks/mock_scorer.h @@ -1,7 +1,7 @@ #include -#include "../scorer.h" -#include "../features/feature.h" +#include "scorer.h" +#include "features/feature.h" namespace extractor { diff --git a/extractor/mocks/mock_suffix_array.h b/extractor/mocks/mock_suffix_array.h index 6886232a..7018acc7 100644 --- a/extractor/mocks/mock_suffix_array.h +++ b/extractor/mocks/mock_suffix_array.h @@ -3,9 +3,9 @@ #include #include -#include "../data_array.h" -#include "../phrase_location.h" -#include "../suffix_array.h" +#include "data_array.h" +#include "phrase_location.h" +#include "suffix_array.h" using namespace std; diff --git a/extractor/mocks/mock_target_phrase_extractor.h b/extractor/mocks/mock_target_phrase_extractor.h index e5e9aeab..6aad853c 100644 --- a/extractor/mocks/mock_target_phrase_extractor.h +++ b/extractor/mocks/mock_target_phrase_extractor.h @@ -1,6 +1,6 @@ #include -#include "../target_phrase_extractor.h" +#include "target_phrase_extractor.h" namespace extractor { diff --git a/extractor/mocks/mock_translation_table.h b/extractor/mocks/mock_translation_table.h index 358c854f..307e4282 100644 --- a/extractor/mocks/mock_translation_table.h +++ b/extractor/mocks/mock_translation_table.h @@ -1,6 +1,6 @@ #include -#include "../translation_table.h" +#include "translation_table.h" namespace extractor { diff --git a/extractor/mocks/mock_vocabulary.h b/extractor/mocks/mock_vocabulary.h index 802c29b4..042c9ce2 100644 --- a/extractor/mocks/mock_vocabulary.h +++ b/extractor/mocks/mock_vocabulary.h @@ -1,6 +1,6 @@ #include -#include "../vocabulary.h" +#include "vocabulary.h" namespace extractor { diff --git a/extractor/precomputation_test.cc b/extractor/precomputation_test.cc index 04e3850d..363febb7 100644 --- a/extractor/precomputation_test.cc +++ b/extractor/precomputation_test.cc @@ -40,7 +40,6 @@ TEST_F(PrecomputationTest, TestCollocations) { Precomputation precomputation(suffix_array, 3, 3, 10, 5, 1, 4, 2); Index collocations = precomputation.GetCollocations(); - EXPECT_EQ(-1, precomputation.NON_TERMINAL); vector key = {2, 3, -1, 2}; vector expected_value = {1, 5, 1, 8, 5, 8, 5, 11, 8, 11}; EXPECT_EQ(expected_value, collocations[key]); @@ -69,33 +68,33 @@ TEST_F(PrecomputationTest, TestCollocations) { expected_value = {1, 6, 1, 9, 5, 9}; EXPECT_EQ(expected_value, collocations[key]); - key = {2, -1, 2, -1, 2}; + key = {2, -1, 2, -2, 2}; expected_value = {1, 5, 8, 5, 8, 11}; EXPECT_EQ(expected_value, collocations[key]); - key = {2, -1, 2, -1, 3}; + key = {2, -1, 2, -2, 3}; expected_value = {1, 5, 9}; EXPECT_EQ(expected_value, collocations[key]); - key = {2, -1, 3, -1, 2}; + key = {2, -1, 3, -2, 2}; expected_value = {1, 6, 8, 5, 9, 11}; EXPECT_EQ(expected_value, collocations[key]); - key = {2, -1, 3, -1, 3}; + key = {2, -1, 3, -2, 3}; expected_value = {1, 6, 9}; EXPECT_EQ(expected_value, collocations[key]); - key = {3, -1, 2, -1, 2}; + key = {3, -1, 2, -2, 2}; expected_value = {2, 5, 8, 2, 5, 11, 2, 8, 11, 6, 8, 11}; EXPECT_EQ(expected_value, collocations[key]); - key = {3, -1, 2, -1, 3}; + key = {3, -1, 2, -2, 3}; expected_value = {2, 5, 9}; EXPECT_EQ(expected_value, collocations[key]); - key = {3, -1, 3, -1, 2}; + key = {3, -1, 3, -2, 2}; expected_value = {2, 6, 8, 2, 6, 11, 2, 9, 11, 6, 9, 11}; EXPECT_EQ(expected_value, collocations[key]); - key = {3, -1, 3, -1, 3}; + key = {3, -1, 3, -2, 3}; expected_value = {2, 6, 9}; EXPECT_EQ(expected_value, collocations[key]); // Exceeds max_rule_symbols. - key = {2, -1, 2, -1, 2, 3}; + key = {2, -1, 2, -2, 2, 3}; EXPECT_EQ(0, collocations.count(key)); // Contains non frequent pattern. key = {2, -1, 5}; -- cgit v1.2.3