Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Suggest to use SimpleImputer for str categorical data with None values #16739

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 3 commits into from
Closed

Suggest to use SimpleImputer for str categorical data with None values #16739

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
1c0f12b
Suggest to use SimpleImputer for str categorical data with None values
ogrisel Mar 21, 2020
6c1c8e5
Update sklearn/preprocessing/_label.py
ogrisel Mar 21, 2020
9aa8e48
PEP8
ogrisel Mar 21, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 16 additions & 6 deletions sklearn/preprocessing/_label.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -109,13 +109,23 @@ def _encode(values, uniques=None, encode=False, check_unknown=True):
"""
if values.dtype == object:
try:
res = _encode_python(values, uniques, encode)
return _encode_python(values, uniques, encode)
except TypeError:
types = sorted(t.__qualname__
for t in set(type(v) for v in values))
raise TypeError("Encoders require their input to be uniformly "
f"strings or numbers. Got {types}")
return res
unique_types = set(type(v) for v in values)
type_names = sorted(t.__qualname__ for t in unique_types)
error_msg = ("Encoders require their input to be uniformly "
f"strings or numbers. Got: {type_names}.")
if unique_types == {str, type(None)}:
# Be extra helpful to the user for this specific yet common
# case:
error_msg += (
'\nPreprocess the features with sklearn.impute.'
'SimpleImputer(strategy="constant", missing_values=None, '
'fill_value="missing") to deal with None-encoded '
'missing values in other-wise string encoded categorical'
'data.'
)
raise TypeError(error_msg)
else:
return _encode_numpy(values, uniques, encode,
check_unknown=check_unknown)
Expand Down
16 changes: 14 additions & 2 deletions sklearn/preprocessing/tests/test_encoders.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -692,7 +692,19 @@ def test_encoders_has_categorical_tags(Encoder):

@pytest.mark.parametrize('Encoder', [OneHotEncoder, OrdinalEncoder])
def test_encoders_does_not_support_none_values(Encoder):
values = [["a"], [1]]
expected_msg = (
r"Encoders require their input to be uniformly strings or numbers\. "
r"Got: \['int', 'str'\]\."
)
with pytest.raises(TypeError, match=expected_msg):
Encoder().fit(values)

# Special case with extra missing value imputation suggestion:
values = [["a"], [None]]
with pytest.raises(TypeError, match="Encoders require their input to be "
"uniformly strings or numbers."):
expected_msg = (
r'SimpleImputer\(strategy="constant", missing_values=None, '
r'fill_value="missing"\)'
)
with pytest.raises(TypeError, match=expected_msg):
Encoder().fit(values)