def test_max_values_mapping(df_cyclical):
cyclical = CyclicalFeatures(variables="day", max_values={"day": 31})
X = cyclical.fit_transform(df_cyclical)
transf_df = df_cyclical.copy()
transf_df["day_sin"] = [
0.937752,
0.988468,
0.848644,
0.571268,
0.201298,
0.394355,
0.724792,
]
transf_df["day_cos"] = [
0.347305,
0.151428,
0.528964,
0.820763,
0.979530,
0.918958,
0.688967,
]
pd.testing.assert_frame_equal(X, transf_df)
def test_general_transformation_dropping_original_variables(df_cyclical):
# test case 1: just one variable, but dropping the variable after transformation
cyclical = CyclicalFeatures(variables=["day"], drop_original=True)
X = cyclical.fit_transform(df_cyclical)
transf_df = df_cyclical.copy()
# expected output
transf_df["day_sin"] = [
-0.78183,
0.0,
-0.97493,
0.43388,
0.78183,
0.97493,
-0.43388,
]
transf_df["day_cos"] = [
0.623490,
1.0,
-0.222521,
-0.900969,
0.623490,
-0.222521,
-0.900969,
]
transf_df = transf_df.drop(columns="day")
# test fit attr
assert cyclical.n_features_in_ == 2
assert cyclical.max_values_ == {"day": 7}
# test transform output
pd.testing.assert_frame_equal(X, transf_df)
def test_general_transformation_without_dropping_variables(df_cyclical):
# test case 1: just one variable.
cyclical = CyclicalFeatures(variables=["day"])
X = cyclical.fit_transform(df_cyclical)
transf_df = df_cyclical.copy()
# expected output
transf_df["day_sin"] = [
-0.78183,
0.0,
-0.97493,
0.43388,
0.78183,
0.97493,
-0.43388,
]
transf_df["day_cos"] = [
0.623490,
1.0,
-0.222521,
-0.900969,
0.623490,
-0.222521,
-0.900969,
]
# fit attr
assert cyclical.max_values_ == {"day": 7}
# test transform output
pd.testing.assert_frame_equal(X, transf_df)
def test_automatically_find_variables(df_cyclical):
# test case 2: automatically select variables
cyclical = CyclicalFeatures(variables=None, drop_original=True)
X = cyclical.fit_transform(df_cyclical)
transf_df = df_cyclical.copy()
# expected output
transf_df["day_sin"] = [
-0.78183,
0.0,
-0.97493,
0.43388,
0.78183,
0.97493,
-0.43388,
]
transf_df["day_cos"] = [
0.62349,
1.0,
-0.222521,
-0.900969,
0.62349,
-0.222521,
-0.900969,
]
transf_df["months_sin"] = [
1.0,
-0.5,
-1.0,
0.0,
0.86603,
0.0,
0.0,
]
transf_df["months_cos"] = [
0.0,
-0.86603,
-0.0,
1.0,
-0.5,
-1.0,
1.0,
]
transf_df = transf_df.drop(columns=["day", "months"])
# test fit attr
assert cyclical.max_values_ == {
"day": 7,
"months": 12,
}
# test transform output
pd.testing.assert_frame_equal(X, transf_df)
def test_raises_error_when_init_parameters_not_permitted(df_cyclical):
with pytest.raises(TypeError):
# when max_values is not a dictionary
CyclicalFeatures(max_values=("dayi", 31))
with pytest.raises(TypeError):
# when max_values values are not integers or string
CyclicalFeatures(max_values={"day": "31"})
with pytest.raises(TypeError):
# when drop original is not a boolean
CyclicalFeatures(drop_original="True")
MathFeatures,
RelativeFeatures,
# FIXME: remove in version 1.4
CombineWithReferenceFeature,
CyclicalTransformer,
MathematicalCombination,
)
from tests.estimator_checks.estimator_checks import check_feature_engine_estimator
_estimators = [
MathFeatures(variables=["0", "1"], func="mean", missing_values="ignore"),
RelativeFeatures(variables=["0", "1"],
reference=["0"],
func=["add"],
missing_values="ignore"),
CyclicalFeatures(),
# FIXME: remove in version 1.4
MathematicalCombination(variables_to_combine=["0", "1"]),
CombineWithReferenceFeature(variables_to_combine=["0"],
reference_variables=["1"]),
CyclicalTransformer(),
]
@pytest.mark.parametrize("estimator", _estimators)
def test_check_estimator_from_sklearn(estimator):
return check_estimator(estimator)
_estimators = [
MathFeatures(variables=["var_1", "var_2", "var_3"], func="mean"),
def test_get_feature_names_out(df_cyclical):
# default features from all variables
transformer = CyclicalFeatures()
X = transformer.fit_transform(df_cyclical)
assert list(X.columns) == transformer.get_feature_names_out()
assert transformer.get_feature_names_out(input_features=["day"]) == [
"day_sin",
"day_cos",
]
assert transformer.get_feature_names_out(
input_features=["day", "months"]) == [
"day_sin",
"day_cos",
"months_sin",
"months_cos",
]
# default features from 1 variable
transformer = CyclicalFeatures(drop_original=True)
X = transformer.fit_transform(df_cyclical)
assert list(X.columns) == transformer.get_feature_names_out()
assert transformer.get_feature_names_out(input_features=["day"]) == [
"day_sin",
"day_cos",
]
assert transformer.get_feature_names_out(
input_features=["day", "months"]) == [
"day_sin",
"day_cos",
"months_sin",
"months_cos",
]
with pytest.raises(ValueError):
# assert error when user passes a string instead of list
transformer.get_feature_names_out(input_features="day")
with pytest.raises(ValueError):
# assert error when uses passes features that were not lagged
transformer.get_feature_names_out(input_features=["color"])
def test_fit_raises_error_if_user_dictionary_key_not_in_df(df_cyclical):
with pytest.raises(KeyError):
transformer = CyclicalFeatures(max_values={"dayi": 31})
transformer.fit(df_cyclical)
def test_fit_raises_error_if_na_in_df(df_na):
# test case 3: when dataset contains na, fit method
with pytest.raises(ValueError):
transformer = CyclicalFeatures()
transformer.fit(df_na)