def prepare_excelcy_data():
excelcy = ExcelCy()
add_stopwords(excelcy.nlp)
excelcy.storage.config = Config(nlp_base='en_core_web_lg',
train_iteration=20,
train_drop=0.2)
excelcy.storage.base_path = str(constants.MODEL_DATA_DIR)
excelcy.storage.source.add(kind='textract',
value='[base_path]/source/training_text.txt')
excelcy.discover()
excelcy.storage.prepare.add(kind='file',
value='[base_path]/prepare/pers.xlsx',
entity='')
excelcy.storage.prepare.add(kind='file',
value='[base_path]/prepare/orgs.xlsx',
entity='')
excelcy.storage.prepare.add(kind='file',
value='[base_path]/prepare/locs.xlsx',
entity='')
excelcy.storage.prepare.add(kind='file',
value='[base_path]/prepare/ships.xlsx',
entity='')
excelcy.storage.prepare.add(kind='file',
value='[base_path]/prepare/misc.xlsx',
entity='')
excelcy.prepare()
excelcy.storage.phase.add('discover')
excelcy.storage.phase.add('prepare')
excelcy.storage.phase.add('train')
excelcy.storage.phase.add('retest')
excelcy.storage.config.prepare_enabled = False
excelcy.save_storage(str(constants.MODEL_DATA_DIR / 'train_model.xlsx'))
def test_readme_03(self):
""" Test: code snippet found in README.rst """
excelcy = ExcelCy()
excelcy.storage.base_path = self.test_data_path
excelcy.storage.config = Config(nlp_base='en_core_web_sm',
train_iteration=2,
train_drop=0.2)
excelcy.storage.source.add(kind='textract',
value='source/source_01.txt')
excelcy.storage.prepare.add(kind='phrase', value='Uber', entity='ORG')
excelcy.discover()
excelcy.prepare()
excelcy.train()
assert excelcy.nlp(
'Uber blew through $1 million a week').ents[0].label_ == 'ORG'