def test_structured_data_input_less_col_name():
(x, _), _1 = utils.dataframe_numpy()
with pytest.raises(ValueError) as info:
adapter = input_adapter.StructuredDataInputAdapter(
column_names=utils.LESS_COLUMN_NAMES_FROM_CSV)
adapter.fit_transform(x)
assert 'Expect column_names to have length' in str(info.value)
def test_structured_data_input_less_col_name_error():
with pytest.raises(ValueError) as info:
adapter = input_adapter.StructuredDataInputAdapter(
column_names=utils.COLUMN_NAMES[:-2])
adapter.fit_transform(pd.read_csv(utils.TRAIN_CSV_PATH))
assert "Expect column_names to have length" in str(info.value)
def test_structured_data_get_col_names():
adapter = input_adapter.StructuredDataInputAdapter()
x = pd.read_csv(utils.TRAIN_CSV_PATH)
x.pop("survived")
adapter.fit_transform(x)
assert adapter.column_names[0] == "sex"
def test_structured_data_infer_col_types():
adapter = input_adapter.StructuredDataInputAdapter()
x = pd.read_csv(utils.TRAIN_CSV_PATH)
x.pop("survived")
adapter.fit_transform(x)
assert adapter.column_types == utils.COLUMN_TYPES
def test_structured_data_col_type_no_name_error():
with pytest.raises(ValueError) as info:
adapter = input_adapter.StructuredDataInputAdapter(
column_types=utils.COLUMN_TYPES)
adapter.transform(
pd.read_csv(utils.TRAIN_CSV_PATH).to_numpy().astype(np.unicode))
assert str(info.value) == "Column names must be specified."
def test_structured_data_input_transform_to_dataset():
x = tf.data.Dataset.from_tensor_slices(
pd.read_csv(utils.TRAIN_CSV_PATH).to_numpy().astype(np.unicode))
adapter = input_adapter.StructuredDataInputAdapter()
x = adapter.fit_transform(x)
assert isinstance(x, tf.data.Dataset)
def test_partial_column_types():
adapter = input_adapter.StructuredDataInputAdapter(
column_names=utils.COLUMN_NAMES_FROM_CSV,
column_types=utils.PARTIAL_COLUMN_TYPES_FROM_CSV)
(x, y), (val_x, val_y) = utils.dataframe_numpy()
dataset = x.values.astype(np.unicode)
adapter.transform(dataset)
assert adapter.column_types['fare'] == 'categorical'
def test_structured_data_input_unsupported_type():
x = 'unknown'
with pytest.raises(TypeError) as info:
adapter = input_adapter.StructuredDataInputAdapter(
column_names=utils.COLUMN_TYPES_FROM_NUMPY,
column_types=utils.COLUMN_TYPES_FROM_NUMPY)
adapter.transform(x)
assert 'Unsupported type' in str(info.value)
def test_structured_data_input_col_type_without_name():
num_data = 500
train_x = utils.generate_structured_data(num_data)
with pytest.raises(ValueError) as info:
adapter = input_adapter.StructuredDataInputAdapter(
column_types=utils.COLUMN_TYPES_FROM_NUMPY)
adapter.transform(train_x)
assert str(info.value) == 'Column names must be specified.'
def test_structured_data_input_unsupported_type_error():
with pytest.raises(TypeError) as info:
adapter = input_adapter.StructuredDataInputAdapter(
column_names=utils.COLUMN_NAMES,
column_types=utils.COLUMN_TYPES,
)
adapter.transform("unknown")
assert "Unsupported type" in str(info.value)
def test_structured_data_input_name_type_mismatch():
(x, _), _1 = utils.dataframe_dataframe()
column_types = copy.copy(utils.COLUMN_TYPES_FROM_CSV)
column_types['age_'] = column_types.pop('age')
with pytest.raises(ValueError) as info:
adapter = input_adapter.StructuredDataInputAdapter(
column_types=column_types)
adapter.transform(x)
assert 'Column_names and column_types are mismatched.' in str(info.value)
def test_structured_data_input_name_type_mismatch_error():
column_types = copy.copy(utils.COLUMN_TYPES)
column_types["age_"] = column_types.pop("age")
with pytest.raises(ValueError) as info:
adapter = input_adapter.StructuredDataInputAdapter(
column_types=column_types)
adapter.transform(pd.read_csv(utils.TRAIN_CSV_PATH))
assert "Column_names and column_types are mismatched." in str(info.value)
def test_dont_infer_specified_column_types():
column_types = copy.copy(utils.COLUMN_TYPES)
column_types.pop("sex")
column_types["age"] = "categorical"
adapter = input_adapter.StructuredDataInputAdapter(
column_names=utils.COLUMN_NAMES,
column_types=column_types,
)
dataset = pd.read_csv(utils.TRAIN_CSV_PATH).to_numpy().astype(np.unicode)
adapter.transform(dataset)
assert adapter.column_types["age"] == "categorical"
def test_structured_data_input_dataset():
(x, _), _1 = utils.dataframe_dataframe()
x = tf.data.Dataset.from_tensor_slices(x.to_numpy().astype(np.unicode))
adapter = input_adapter.StructuredDataInputAdapter()
x = adapter.fit_transform(x)
assert isinstance(x, tf.data.Dataset)
def test_structured_data_input_transform():
(x, _), _1 = utils.dataframe_dataframe()
adapter = input_adapter.StructuredDataInputAdapter()
adapter.fit_transform(x)
assert adapter.column_names[0] == 'sex'
assert adapter.column_types == utils.COLUMN_TYPES_FROM_CSV
