def test_numerical_map_attribute_created(self, data):
"""Tests that the numerical_map attribute is created for the Masker object"""
test_masker = Masker()
test_masker.get_numerical_map(X=data)
assert hasattr(test_masker, "numerical_map")
def test_X_type(self):
"""Test error is thrown if X is not the correct type."""
test_masker = Masker()
with pytest.raises(TypeError):
test_masker.fit(X=123)
def test_categorical_map_attribute_created(self, data):
"""Test that the categorical_map attribute is created for the Masker object."""
test_masker = Masker()
test_masker.get_categorical_map(X=data)
assert hasattr(test_masker, "categorical_map")
def test_X_type(self):
"""Test error is thrown if X is not a pd.DataFrame."""
test_masker = Masker()
with pytest.raises(TypeError):
test_masker.get_column_map(X=123)
def test_columns_multiple_columns(self):
"""Test column names mapped correctly for multiple columns."""
dummy_data = pd.DataFrame({"col1": [1, 2, 3], "col2": [2, 3, 4]})
test_masker = Masker()
dummy_data = test_masker.fit_transform(dummy_data)
assert all(dummy_data.columns == ["column_0", "column_1"])
def test_columns_single_column(self):
"""Test column names mapped correctly for a single column."""
dummy_data = pd.DataFrame({"col1": [1, 2, 3]})
test_masker = Masker()
dummy_data = test_masker.fit_transform(dummy_data)
assert dummy_data.columns[0] == "column_0"
def test_values_single_column(self):
"""Test column values mapped correctly for a single column."""
dummy_data = pd.DataFrame({"col1": ["A", "B", "A"]})
test_masker = Masker()
dummy_data = test_masker.fit_transform(dummy_data)
assert all(dummy_data.iloc[:, 0] == ["level_0", "level_1", "level_0"])
def test_numerical_map_values(self, data):
"""Test map values for MinMaxScaler."""
test_masker = Masker()
test_masker.get_numerical_map(X=data)
assert list(test_masker.numerical_map.keys()) == ["col2"]
assert isinstance(test_masker.numerical_map["col2"], MinMaxScaler)
def test_get_column_map_called(self, data, mocker):
"""Test that the get_column_map method is called."""
mocker.patch.object(Masker, "get_column_map")
test_masker = Masker()
test_masker.fit(X=data)
Masker.get_column_map.assert_called()
def test_values_multiple_columns(self):
"""Test column values maped correctly for multple columns."""
dummy_data = pd.DataFrame({
"col1": ["A", "B", "C"],
"col2": ["1", "1", "1"]
})
test_masker = Masker()
dummy_data = test_masker.fit_transform(dummy_data)
assert all(dummy_data.iloc[:, 0] == ["level_0", "level_1", "level_2"])
assert all(dummy_data.iloc[:, 1] == ["level_0", "level_0", "level_0"])
def test_column_map_attribute_value(self, data):
"""Test that the column_map attribute takes the expect values."""
test_masker = Masker()
test_masker.get_column_map(X=data)
expected_map = {
"col1": "column_0",
"col2": "column_1",
"col3": "column_2",
"col4": "column_3",
}
assert test_masker.column_map == expected_map
def test_dataframe_matches_with_scaling(self):
"""Full test that the dataframe is as expected, with numeric scaling."""
dummy_data = pd.DataFrame({
"col1": ["1", "2", "1"],
"col2": [1, 2, 3],
"col3": ["red", "red", "blue"]
})
expected_data = pd.DataFrame({
"column_0": ["level_0", "level_1", "level_0"],
"column_1": [0.0, 0.5, 1.0],
"column_2": ["level_0", "level_0", "level_1"],
})
test_masker = Masker(numerical_scaling=True)
dummy_data = test_masker.fit_transform(dummy_data)
assert dummy_data.equals(expected_data)
def test_categorical_map_value(self, data):
"""Test that the categorical_map attribute takes the correct value."""
test_masker = Masker()
test_masker.get_categorical_map(X=data)
expected_map = {
"col1": {
"one": "level_0",
"two": "level_1",
"three": "level_2"
},
"col3": {
"1": "level_0",
"2": "level_1"
},
"col4": {
np.NaN: "level_0",
"blue": "level_1"
},
}
assert test_masker.categorical_map == expected_map
def test_numerical_scaling_attribute_created(self):
"""Tests the Masker object is created with the numerical_scaling attribute."""
test_masker = Masker()
assert hasattr(test_masker, "numerical_scaling")
def test_numerical_scaling_type(self):
"""Test error is thrown if numerical_scaling is not the correct type."""
with pytest.raises(TypeError):
test_masker = Masker(numerical_scaling=123)
def test_scaling_method_default(self):
"""Test scaling_method argument defaults correctly."""
test_masker = Masker()
assert test_masker.scaling_method == "MinMaxScaler"
def test_scaling_method_attribute_created(self):
"""Test the Masker object is created with the scaling_method attribute."""
test_masker = Masker()
assert hasattr(test_masker, "scaling_method")
def test_scaling_method_value(self):
"""Test error is thrown if scaling_method is not the correct value."""
with pytest.raises(ValueError):
test_masker = Masker(scaling_method="dummy_method")
def test_scaling_method_type(self):
"""Test error is thrown if scaling_method is not the correct type."""
with pytest.raises(TypeError):
test_masker = Masker(scaling_method=123)
def test_numerical_scaling_default(self):
"""Test numerical_scaling argument defaults correctly."""
test_masker = Masker()
assert test_masker.numerical_scaling
