def create_and_fill_variant_collection(frames_collection):
frame_variant_collection = FrameVariantsCollection()
frame_variant_collection.fill_from_iterable(
variants_with_id=frames_collection.iter_frame_id_and_variants(),
overwrite_existed_variant=True,
raise_error_on_existed_variant=False)
return frame_variant_collection
def __iter_frame_variants():
frames_collection = RuSentiFramesCollection.read_collection(RuSentiFramesVersions.V20)
frame_variants = FrameVariantsCollection()
frame_variants.fill_from_iterable(variants_with_id=frames_collection.iter_frame_id_and_variants(),
overwrite_existed_variant=True,
raise_error_on_existed_variant=False)
for v, _ in frame_variants.iter_variants():
yield v
def __create_frames_variants_collection(self):
frames = RuSentiFramesCollection.read_collection(
RuSentiFramesVersions.V20)
frame_variant_collection = FrameVariantsCollection()
frame_variant_collection.fill_from_iterable(
variants_with_id=frames.iter_frame_id_and_variants(),
overwrite_existed_variant=True,
raise_error_on_existed_variant=False)
return frame_variant_collection
def test_reading(self):
# Initializing logger
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)
logging.basicConfig(level=logging.DEBUG)
frames = RuSentiFramesCollection.read_collection(
RuSentiFramesVersions.V20)
for frame_id in frames.iter_frames_ids():
# id
logger.info("Frame: {}".format(frame_id))
# titles
logger.info("Titles: {}".format(",".join(
frames.get_frame_titles(frame_id))))
# variants
logger.info("Variants: {}".format(",".join(
frames.get_frame_variants(frame_id))))
# roles
for role in frames.get_frame_roles(frame_id):
logger.info("Role: {}".format(" -- ".join(
[role.Source, role.Description])))
# states
for state in frames.get_frame_states(frame_id):
logger.info("State: {}".format(",".join(
[state.Role,
state.Label.to_class_str(),
str(state.Prob)])))
# polarity
for polarity in frames.get_frame_polarities(frame_id):
logger.info("Polarity: {}".format(",".join([
polarity.Source, polarity.Destination,
polarity.Label.to_class_str()
])))
has_a0_a1_pol = frames.try_get_frame_polarity(frame_id,
role_src="a0",
role_dest="a1")
logger.info(
"Has a0->a1 polarity: {}".format(has_a0_a1_pol is not None))
# frame variants.
frame_variants = FrameVariantsCollection()
frame_variants.fill_from_iterable(
variants_with_id=frames.iter_frame_id_and_variants(),
overwrite_existed_variant=True,
raise_error_on_existed_variant=False)
frame_variant = frame_variants.get_variant_by_value("хвалить")
assert (isinstance(frame_variant, FrameVariant))
logger.info("FrameVariantValue: {}".format(frame_variant.get_value()))
logger.info("FrameID: {}".format(frame_variant.FrameID))
def test_parsing(self):
# Initializing logger.
logger = logging.getLogger(__name__)
logger.setLevel(logging.DEBUG)
logging.basicConfig(level=logging.DEBUG)
# Initializing stemmer.
stemmer = MystemWrapper()
# frame and variants.
frames = RuSentiFramesCollection.read_collection(version=RuSentiFramesVersions.V20)
frame_variants = FrameVariantsCollection()
frame_variants.fill_from_iterable(variants_with_id=frames.iter_frame_id_and_variants(),
overwrite_existed_variant=True,
raise_error_on_existed_variant=False)
text_parser = BaseTextParser(pipeline=[RuSentRelTextEntitiesParser(),
DefaultTextTokenizer(keep_tokens=True),
LemmasBasedFrameVariantsParser(frame_variants=frame_variants,
stemmer=stemmer,
save_lemmas=False),
FrameVariantsSentimentNegation()])
# Reading synonyms collection.
synonyms = RuSentRelSynonymsCollectionProvider.load_collection(stemmer=stemmer)
version = RuSentRelVersions.V11
for doc_id in RuSentRelIOUtils.iter_collection_indices(version):
# Parsing
news = RuSentRelNews.read_document(doc_id=doc_id,
synonyms=synonyms,
version=version)
# Perform text parsing.
parsed_news = NewsParser.parse(news=news, text_parser=text_parser)
debug_show_news_terms(parsed_news=parsed_news)
def setUpClass(cls):
cls.stemmer = MystemWrapper()
cls.entities_formatter = RussianEntitiesCasedFormatter(
pos_tagger=POSMystemWrapper(Mystem(entire_input=False)))
cls.synonyms = RuSentRelSynonymsCollectionProvider.load_collection(
stemmer=cls.stemmer)
cls.frames_collection = RuSentiFramesCollection.read_collection(
version=RuSentiFramesVersions.V10)
cls.unique_frame_variants = FrameVariantsCollection()
cls.unique_frame_variants.fill_from_iterable(
variants_with_id=cls.frames_collection.iter_frame_id_and_variants(
),
overwrite_existed_variant=True,
raise_error_on_existed_variant=False)
def __about(frames_collection, pos_tagger):
all_frame_entries = list(frames_collection.iter_frame_id_and_variants())
unique_frame_variants = FrameVariantsCollection()
unique_frame_variants.fill_from_iterable(
variants_with_id=all_frame_entries,
overwrite_existed_variant=True,
raise_error_on_existed_variant=False)
assert (isinstance(frames_collection, RuSentiFramesCollection))
unique_variants = list(unique_frame_variants.iter_variants())
phrases = []
nouns = []
verbs = []
other = []
for frame_id, variant in unique_variants:
assert (isinstance(variant, FrameVariant))
terms = list(variant.iter_terms())
if len(terms) > 1:
phrases.append(variant.get_value())
continue
pos_type = pos_tagger.get_term_pos(terms[0])
if pos_tagger.is_noun(pos_type):
nouns.append(terms[0])
continue
if pos_tagger.is_verb(pos_type):
verbs.append(terms[0])
continue
other.append(terms[0])
titles = []
for frame_id in frames_collection.iter_frames_ids():
titles.extend(frames_collection.get_frame_titles(frame_id))
print("Frames count:", len(list(frames_collection.iter_frames_ids())))
print("---------------")
print()
print("Quantitative characteristics of the RuSentiFrames entries:")
print("Verbs:", len(verbs))
print("Nouns:", len(nouns))
print("Phrases:", len(phrases))
print("Other:", len(other))
print("Unique entries:", len(unique_variants))
print("Total entries: ", len(all_frame_entries))
print()
print(
"The distribution of RuSentiFrames text entries according to attitudes:"
)
print(
"A0 to A1 Pos",
len(
__get_variants_with_polarities(frames_collection=frames_collection,
role_src='a0',
role_dest='a1',
label=PositiveLabel())))
print(
"A0 to A1 Neg",
len(
__get_variants_with_polarities(frames_collection=frames_collection,
role_src='a0',
role_dest='a1',
label=NegativeLabel())))
print(
"Author to A0 Pos",
len(
__get_variants_with_polarities(frames_collection=frames_collection,
role_src='author',
role_dest='a0',
label=PositiveLabel())))
print(
"Author to A0 Neg",
len(
__get_variants_with_polarities(frames_collection=frames_collection,
role_src='author',
role_dest='a0',
label=NegativeLabel())))
print(
"Author to A1 Pos",
len(
__get_variants_with_polarities(frames_collection=frames_collection,
role_src='author',
role_dest='a1',
label=PositiveLabel())))
print(
"Author to A1 Neg",
len(
__get_variants_with_polarities(frames_collection=frames_collection,
role_src='author',
role_dest='a1',
label=NegativeLabel())))
print()
print(
"The distribution of RuSentiFrames text entries according to effects on main participants:"
)
print(
"A0 Pos",
len(
__get_frame_effects(frames_collection=frames_collection,
role='a0',
label=PositiveLabel())))
print(
"A0 Neg",
len(
__get_frame_effects(frames_collection=frames_collection,
role='a0',
label=NegativeLabel())))
print(
"A1 Pos",
len(
__get_frame_effects(frames_collection=frames_collection,
role='a1',
label=PositiveLabel())))
print(
"A1 Neg",
len(
__get_frame_effects(frames_collection=frames_collection,
role='a1',
label=NegativeLabel())))