def __init__(self,
max_left_length=10,
max_right_length=9,
predictor_headers=None,
verbose=True,
min_num_of_features=1,
max_frequency_when_noun_is_eojeol=30,
eojeol_counter_filtering_checkpoint=200000,
min_eojeol_frequency=1,
extract_compound=True,
extract_pos_feature=False,
extract_determiner=False,
postprocessing=None,
logpath=None):
self.max_left_length = max_left_length
self.max_right_length = max_right_length
self.lrgraph = None
self.verbose = verbose
self.min_num_of_features = min_num_of_features
self.max_frequency_when_noun_is_eojeol = max_frequency_when_noun_is_eojeol
self.eojeol_counter_filtering_checkpoint = eojeol_counter_filtering_checkpoint
self.min_eojeol_frequency = min_eojeol_frequency
self.extract_compound = extract_compound
self.extract_pos_feature = extract_pos_feature
self.extract_determiner = extract_determiner
self.logpath = logpath
if logpath:
check_dirs(logpath)
if not postprocessing:
postprocessing = [
'detaching_features', 'ignore_features', 'ignore_NJ'
]
elif isinstance(postprocessing) == str:
postprocessing = [postprocessing]
self.postprocessing = postprocessing
if not predictor_headers:
predictor_headers = self._set_default_predictor_header()
self._load_predictor(predictor_headers)
def write_log(path, header, words):
check_dirs(path)
with open(path, 'a', encoding='utf-8') as f:
f.write('\n{}\n'.format(header))
for word in sorted(words):
f.write('{}\n'.format(word))