def test_write_filled_tsv_taxonomy(self):
temp_file = declare_temp_file('.tsv')
# create objects from the dict (as expected by the io.py module)
objects = {}
for key in taxonomy_dict:
objects[key] = Taxonomy(**taxonomy_dict[key])
# then write created objects
io.write(objects, temp_file)
output = io.read(temp_file)
for key in output:
self.assertDictEqual(output[key].__dict__, taxonomy_dict[key])
def _read_table(self, path):
"""
Read in a single table.
- Parse taxonomies
- Create a CodeTable instance
- Assign taxonomies to the new CodeTable
params:
path (str)
"""
parsed_taxonomies = io.read(path)
table = CodeTable()
table.read(parsed_taxonomies)
return table
def parse_config(self, file_path):
"""
Read and parse the config file.
params:
file_path (str): path to config file
returns:
dict{str:list[str]}: parsed config as a dict
"""
self.parser = io.read(file_path)
for option in self.options:
if option.required:
self._parse_required(option)
else:
self._parse_optional(option)
def train_vector_space(self, train_params, output_path):
"""
Train vector space model.
params:
train_params (dict)
"""
if io.exists(output_path):
logger.debug('Vector space already exists, loading %s' %
output_path)
self.vector_space_model = io.read(output_path)
else:
logger.info('Training vector space using word2vec')
self._collect_phrases()
phrases_as_tokens = self._phrases_to_tokens()
self._start_word2vec(phrases_as_tokens, train_params)
self._assign_vectors_to_phrases()
logger.debug('Writing vector space to %s' %
output_path)
io.write(self.vector_space_model, output_path)
def test_read_empty_tsv_taxonomy(self):
temp_file = create_temp_file(code_table_tsv, '.tsv')
output = io.read(temp_file)
for key in output:
self.assertDictEqual(output[key].__dict__, code_table_dict[key])
def test_read_filled_xml_taxonomy(self):
temp_file = create_temp_file(taxonomy_xml, '.xml')
output = io.read(temp_file)
for key in output:
self.assertDictEqual(output[key].__dict__, taxonomy_dict[key])
def test_read_wrong_extension(self):
temp_file = create_temp_file(taxonomy_tsv, '.json')
io.read(temp_file)