def test_train_and_evaluate(self):
data = pandas.read_csv(str(TEST_DATA_DIR / "test_data.csv.xz"), index_col=0,
keep_default_na=False)
vocabulary_file = str(TEST_DATA_DIR / "test_frequencies.csv.xz")
model = train_and_evaluate(data, data, vocabulary_file, vocabulary_file,
str(TEST_DATA_DIR / "test_ft.bin"))
suggestions = model.suggest_on_file(str(TEST_DATA_DIR / "test_data.csv.xz"))
self.assertSetEqual(set(suggestions.keys()), set(data.index))
def cli_train_corrector(train: str, test: str, vocabulary_path: str,
frequencies_path: str, fasttext_path: str,
corrector_path: str) -> None:
"""Entry point for `train_and_evaluate`."""
train = pandas.read_csv(train, index_col=0, keep_default_na=False)
test = pandas.read_csv(test, index_col=0, keep_default_na=False)
model = train_and_evaluate(train, test, vocabulary_path, frequencies_path,
fasttext_path)
model.save(corrector_path, series=0.0)
def cli_train_corrector(train: str, test: str, vocabulary_path: str,
frequencies_path: str, fasttext_path: str,
corrector_path: str, config: Mapping[str,
Any]) -> None:
"""Entry point for `train_and_evaluate`."""
train = pandas.read_csv(train, index_col=0, keep_default_na=False)
test = pandas.read_csv(test, index_col=0, keep_default_na=False)
model = train_and_evaluate(train, test, vocabulary_path, frequencies_path,
fasttext_path, config.get("generation", {}),
config.get("ranking", {}))
model.save(corrector_path, series=0.0)