def get_names(self, list_) -> list:
names = []
utils = Utils()
stringList = []
for splitchunk in list_:
if not utils.isValidName(splitchunk.name):
continue
stringList.append(splitchunk.name)
content = ' '.join(stringList)
config = configparser.ConfigParser()
config.read('settings.ini')
opennlp_dir = config['options']['opennlp_dir']
models_dir = config['options']['models_dir']
language = 'en'
tt = OpenNLPTagger(language=language,
path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_model=os.path.join(models_dir,
'en-pos-maxent.bin'))
phrase = str(content)
sentence = tt.tag(phrase)
return self.getValidNames(sentence)
def test_opennlp_tagger():
dirname, filename = os.path.split(os.path.abspath(__file__))
language = 'en'
tt = OpenNLPTagger(language=language,
path_to_bin=os.path.join(dirname, 'apache-opennlp', 'bin'),
path_to_model=os.path.join(dirname, 'opennlp_models', 'en-pos-maxent.bin'))
phrase = 'Pierre Vinken , 61 years old , will join the board as a nonexecutive director Nov. 29 .'
en_tags = tt.tag(phrase)
assert en_tags[0][0] == 'Pierre'
assert en_tags[0][1] == 'NNP'
def test_opennlp_tagger(self):
language = 'en'
tt = OpenNLPTagger(language=language,
path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_model=os.path.join(models_dir,
'en-pos-maxent.bin'))
phrase = 'Pierre Vinken , 61 years old , will join the board as a nonexecutive director Nov. 29 .'
en_tags = tt.tag(phrase)
print(en_tags)
assert en_tags[0][0] == 'Pierre'
assert en_tags[0][1] == 'NNP'
def test_opennlp_chunker(self):
language = 'en'
tt = OpenNLPTagger(language=language,
path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_model=os.path.join(models_dir,
'en-pos-maxent.bin'))
phrase = 'Pierre Vinken , 61 years old , will join the board as a nonexecutive director Nov. 29 .'
sentence = tt.tag(phrase)
cp = OpenNLPChunker(path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_chunker=os.path.join(
models_dir, 'en-chunker.bin'))
print(cp.parse(sentence))
def opennlp_test(self, content):
config = configparser.ConfigParser()
config.read('settings.ini')
opennlp_dir = config['options']['opennlp_dir']
models_dir = config['options']['models_dir']
language = 'en'
tt = OpenNLPTagger(language=language,
path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_model=os.path.join(models_dir,
'en-pos-maxent.bin'))
phrase = str(content)
sentence = tt.tag(phrase)
return self.getValidNames(sentence)
def test_opennlp_ner_chunker_bracketed(self):
language = 'en'
tt = OpenNLPTagger(language=language,
path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_model=os.path.join(models_dir,
'en-pos-maxent.bin'))
phrase = 'Pierre Vinken , ( 61 years old ) , will join Martin Vinken as a nonexecutive director Nov. 29 .'
sentence = tt.tag(phrase)
cp = OpenNERChunker(path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_chunker=os.path.join(
models_dir, '{}-chunker.bin'.format(language)),
path_to_ner_model=os.path.join(
models_dir,
'{}-ner-person.bin'.format(language)))
print(cp.parse(sentence))
def test_opennlp_chunker_de(self):
language = 'de'
tt = OpenNLPTagger(language=language,
path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_model=os.path.join(models_dir,
'de-pos-maxent.bin'))
phrase = 'Das Haus hat einen großen hübschen Garten.'
sentence = tt.tag(phrase)
print(sentence)
# There should not be OpenNLP chunker for German language, thus OSError is thrown in Linux
if os.name != 'nt':
with self.assertRaises(OSError):
cp = OpenNLPChunker(
path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_chunker=os.path.join(models_dir, 'de-chunker.bin'))
print(cp.parse(sentence))
def test_opennlp_tagger_list(self):
language = 'en'
tt = OpenNLPTagger(language=language,
path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_model=os.path.join(models_dir,
'en-pos-maxent.bin'))
phrase = [
'Pierre', 'Vinken'
','
'61', 'years', 'old', ',', 'will', 'join', 'the', 'board', 'as',
'a', 'nonexecutive', 'director', 'Nov.', '29', '.'
]
en_tags = tt.tag(phrase)
print(en_tags)
assert en_tags[0][0] == 'Pierre'
assert en_tags[0][1] == 'NNP'
def get_maxent_postagger(self, language='en'):
if not self.__has_key(self.__maxent_postagger, language):
dirname, filename = os.path.split(os.path.abspath(__file__))
path_to_model = os.path.join(dirname, 'opennlp_models',
'{}-pos-maxent.bin'.format(language))
if not (os.path.exists(path_to_model)
and os.path.isfile(path_to_model)):
return None
self.__maxent_postagger[language] = OpenNLPTagger(
path_to_bin=os.path.join(dirname, 'apache-opennlp', 'bin'),
path_to_model=path_to_model)
return self.__maxent_postagger[language]
def test_opennlp_ner_multichunker(self):
language = 'en'
tt = OpenNLPTagger(language=language,
path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_model=os.path.join(models_dir,
'en-pos-maxent.bin'))
phrase = 'John Haddock , 32 years old male , travelled to Cambridge , USA in October 20 while paying 6.50 dollars for the ticket'
sentence = tt.tag(phrase)
cp = OpenNERChunkerMulti(
path_to_bin=os.path.join(opennlp_dir, 'bin'),
path_to_chunker=os.path.join(models_dir,
'{}-chunker.bin'.format(language)),
ner_models=[
os.path.join(models_dir, '{}-ner-person.bin'.format(language)),
os.path.join(models_dir, '{}-ner-date.bin'.format(language)),
os.path.join(models_dir,
'{}-ner-location.bin'.format(language)),
os.path.join(models_dir, '{}-ner-time.bin'.format(language)),
os.path.join(models_dir, '{}-ner-money.bin'.format(language))
])
print(cp.parse(sentence))