def __init__(self, config_fp, distance_mode):
Extractor.__init__(self, config_fp)
self.feature_name += '_%s' % distance_mode
self.valid_distance_mode = ['edit_dist', 'compression_dist']
assert distance_mode in self.valid_distance_mode, "Wrong aggregation_mode: %s" % distance_mode
self.distance_mode = distance_mode
self.distance_func = getattr(DistanceUtil, self.distance_mode)
def __init__(self):
self.__featureNames = sorted(
[name for (name, re) in self.DIRECT_FEATS]
+ [name for (name, re) in self.LEMMA_FEATS]
+ self.CALCULATED_FEATS
)
Extractor.__init__(self)
def __init__(self):
Extractor.__init__(self)
self.stop = [
line.strip()
for line in open(PropertiesUtil.getProperty('PATH', 'stopword_fp'),
'r').readlines()
]
self.model = None
def __init__(self, config_fp):
Extractor.__init__(self, config_fp)
powerful_word_fp = '%s/%s.txt' % (self.config.get(
'DIRECTORY',
'devel_pt'), self.config.get('FILE_NAME', 'powerful_word_name'))
self.pword_dict = dict(
PowerfulWord.load_powerful_word(powerful_word_fp))
def __init__(self, config_fp, thresh_num=0, thresh_rate=0.0):
Extractor.__init__(self, config_fp)
powerful_word_fp = '%s/%s.txt' % (self.config.get(
'DIRECTORY',
'devel_pt'), self.config.get('FILE_NAME', 'powerful_word_name'))
self.pword = PowerfulWord.load_powerful_word(powerful_word_fp)
self.pword_oside = PowerfulWordOneSide.init_powerful_word_oside(
self.pword, thresh_num, thresh_rate)
def __init__(self,
config_fp,
n_components=50,
tf_idf=None,
tf_idf_result=None):
Extractor.__init__(self, config_fp)
self.n_components = n_components
self.tf_idf, self.tf_idf_result = tf_idf, tf_idf_result
self.svd_model = self.init_svd()
def __init__(self, config_fp):
Extractor.__init__(self, config_fp)
words_pt = '%s/%s' % (self.config.get('DIRECTORY', 'source_pt'),
self.config.get('FILE_NAME', 'words_txt'))
self.words_dict = self.getWordsDict(words_pt=words_pt)
print("get word2vec_dict...")
word2vec_pt = '%s/%s' % (self.config.get('DIRECTORY', 'source_pt'),
self.config.get('FILE_NAME', 'wiki_es_vec'))
self.word2vec_dict = self.getWord2VecDict(word2vec_pt)
print("get word2vec_dict done")
def __init__(self, config_fp):
Extractor.__init__(self, config_fp)
#words_pt = '%s/%s' % (self.config.get('DIRECTORY', 'source_pt'), self.config.get('FILE_NAME', 'words_txt'))
#self.words_dict = self.getWordsDict(words_pt=words_pt)
train_data = pd.read_csv(
'%s/%s' %
(self.config.get('DIRECTORY', 'csv_spanish_cleaning_pt'),
self.config.get('FILE_NAME',
'preprocessing_train_merge_csv'))).fillna(
value="")
test_data = pd.read_csv(
'%s/%s' %
(self.config.get('DIRECTORY', 'csv_spanish_cleaning_pt'),
self.config.get('FILE_NAME', 'preprocessing_test_csv'))).fillna(
value="")
self.idf = TFIDFWordMatchShare.init_idf(train_data)
def __init__(self, config):
Extractor.__init__(self, config)
self.conf = ConfigParser.ConfigParser()
self.conf.read(config)
train_data = pd.read_csv(
'%s/%s' %
(self.config.get('DIRECTORY', 'csv_spanish_cleaning_pt'),
self.config.get('FILE_NAME',
'preprocessing_train_merge_csv'))).fillna(
value="")
test_data = pd.read_csv(
'%s/%s' %
(self.config.get('DIRECTORY', 'csv_spanish_cleaning_pt'),
self.config.get('FILE_NAME', 'preprocessing_test_csv'))).fillna(
value="")
self.idf = TFIDFWordMatchShare.init_idf(train_data)
def __init__(self, config_fp):
Extractor.__init__(self, config_fp)
#self.graph_result = Graph(config_fp).buildGraph()
self.graph_result = [[['impuestos']],
[['Cómo'],
[
'reporto', 'enviar', 'informar', 'reportar',
'informo'
], ['proveedor']], [['hacer', 'Cómo'],
['pedido']], [['bancaria']],
[['Quiero'], ['pagar']],
[['no', 'ni', 'nunca'], ['pedido']],
[['Donde'], ['cupones']],
[['número'], ['teléfono']],
[['Recibí'], ['pedido']],
[['recibí', 'recibido'], ['no', 'ni', 'nunca']],
[['confiable'], ['vendedor', 'proveedor']],
[['protección'], ['comprador', 'compra']],
[['mi'], ['pregunta']]]
def __init__(self, config_fp, language="es"):
Extractor.__init__(self, config_fp)
self.language = language
def __init__(self, config_fp, n_components=50, tf=None, tf_result=None):
# Use tf (raw term count) features for LDA.
Extractor.__init__(self, config_fp)
self.n_components = n_components
self.tf, self.tf_result = tf, tf_result
self.lda_model = self.init_lda()
def __init__(self):
self.__featureNames = [name for (name,re) in self.DIRECT_FEATS] + self.CALCULATED_FEATS
Extractor.__init__(self)
def __init__(self, config_fp):
Extractor.__init__(self, config_fp)
self.dul_num = self.generate_dul_num()
def __init__(self, config_fp):
Extractor.__init__(self, config_fp)
self.snowball_stemmer = SnowballStemmer('spanish')
def __init__(self, config_fp):
Extractor.__init__(self, config_fp)
self.tfidf, self.tfidf_result = self.init_tfidf()
def __init__(self, config):
Extractor.__init__(self, config)
self.conf = ConfigParser.ConfigParser()
self.conf.read(config)
self.tf, self.tf_result = self.init_tf()
def __init__(self, config_fp, language="en"):
Extractor.__init__(self, config_fp)
# should be modified
self.language = language
def __init__(self):
Extractor.__init__(self)
def __init__(self):
Extractor.__init__(self)
def __init__(self, config, jar_path, max_path_length, max_path_width,
path_dict_and_name):
Extractor.__init__(self, config, jar_path, max_path_length,
max_path_width)
self.word_to_count, self.path_to_count, _, _ = preprocess_test_batch.load_dictionaries(
path_dict_and_name)
def __init__(self):
self.__featureNames = utils.getStatNames("SENT_LENGTH")
Extractor.__init__(self)
