def test_grouping_column_not_in_df(penguins_df):
meta = GroupedTransformer(StandardScaler(),
groups=["island", "unexisting_column"])
# This should raise ValueError
with pytest.raises(ValueError):
meta.fit_transform(penguins_df)
def test_grouping_column_not_in_array(penguins):
X = penguins
meta = GroupedTransformer(StandardScaler(), groups=[0, 5])
# This should raise ValueError
with pytest.raises(ValueError):
meta.fit_transform(X[:, :3])
def test_df_missing_group(penguins_df):
meta = GroupedTransformer(StandardScaler(), groups=["island", "sex"])
# Otherwise the fixture is changed
X = penguins_df.copy()
X.loc[0, "island"] = None
with pytest.raises(ValueError):
meta.fit_transform(X)
def test_array_with_strings():
X = np.array([
("group0", 2),
("group0", 0),
("group1", 1),
("group1", 3),
],
dtype='object')
trf = MinMaxScaler()
transformer = GroupedTransformer(trf, groups=0, use_global_model=False)
transformer.fit_transform(X)
def test_df(penguins_df):
meta = GroupedTransformer(StandardScaler(), groups=["island", "sex"])
transformed = meta.fit_transform(penguins_df)
# 2 columns for grouping not in the result
assert transformed.shape == (penguins_df.shape[0],
penguins_df.shape[1] - 2)
def test_array_with_multiple_string_cols(penguins):
X = penguins
meta = GroupedTransformer(StandardScaler(), groups=[0, -1])
transformed = meta.fit_transform(X)
# 2 columns for grouping not in the result
assert transformed.shape == (X.shape[0], X.shape[1] - 2)
def test_no_grouping(penguins_df):
penguins_numeric = penguins_df[[
"bill_length_mm", "bill_depth_mm", "flipper_length_mm", "body_mass_g"
]]
meta = GroupedTransformer(StandardScaler(), groups=None)
nonmeta = StandardScaler()
assert (meta.fit_transform(penguins_numeric) == nonmeta.fit_transform(
penguins_numeric)).all()
def test_with_y(penguins_df):
X = penguins_df.drop(columns=["sex"])
y = penguins_df["sex"]
meta = GroupedTransformer(StandardScaler(), groups="island")
# This should work fine
transformed = meta.fit_transform(X, y)
# 1 column for grouping not in the result
assert transformed.shape == (X.shape[0], X.shape[1] - 1)