Skip to content

FIX check_estimator fails when validating SGDClassifier with log_loss #24071

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

Merged
merged 7 commits into from
Oct 25, 2022
Merged

FIX check_estimator fails when validating SGDClassifier with log_loss #24071

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
b019fdd
fix check_decision_proba_consistency
MaxwellLZH Aug 1, 2022
97f4ba9
fix linting error
MaxwellLZH Aug 1, 2022
3fbdf24
improve readability
MaxwellLZH Oct 18, 2022
6ab2187
add non-regression test
MaxwellLZH Oct 18, 2022
dda080d
resolve merge conflict
MaxwellLZH Oct 18, 2022
c3e3360
Merge remote-tracking branch 'upstream/main' into pr/24071
thomasjpfan Oct 25, 2022
b1d99a5
DOC Adjust grammar
thomasjpfan Oct 25, 2022
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
16 changes: 15 additions & 1 deletion sklearn/utils/estimator_checks.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3504,7 +3504,21 @@ def check_decision_proba_consistency(name, estimator_orig):
# inversions in case of machine level differences.
a = estimator.predict_proba(X_test)[:, 1].round(decimals=10)
b = estimator.decision_function(X_test).round(decimals=10)
assert_array_equal(rankdata(a), rankdata(b))

rank_proba, rank_score = rankdata(a), rankdata(b)
try:
assert_array_almost_equal(rank_proba, rank_score)
except AssertionError:
# Sometimes, the rounding applied on the probabilities will have
# ties that are not present in the scores because it is
# numerically more precise. In this case, we relax the test by
# grouping the decision function scores based on the probability
# rank and check that the score is monotonically increasing.
grouped_y_score = np.array(
[b[rank_proba == group].mean() for group in np.unique(rank_proba)]
)
sorted_idx = np.argsort(grouped_y_score)
assert_array_equal(sorted_idx, np.arange(len(sorted_idx)))


def check_outliers_fit_predict(name, estimator_orig):
Expand Down
11 changes: 11 additions & 0 deletions sklearn/utils/tests/test_estimator_checks.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,6 +36,7 @@
from sklearn.utils import all_estimators
from sklearn.exceptions import SkipTestWarning
from sklearn.utils.metaestimators import available_if
from sklearn.utils.estimator_checks import check_decision_proba_consistency
from sklearn.utils._param_validation import Interval, StrOptions

from sklearn.utils.estimator_checks import (
Expand Down Expand Up @@ -1159,3 +1160,13 @@ class OutlierDetectorWithConstraint(OutlierDetectorWithoutConstraint):
detector = OutlierDetectorWithConstraint()
with raises(AssertionError, match=err_msg):
check_outlier_contamination(detector.__class__.__name__, detector)


def test_decision_proba_tie_ranking():
"""Check that in case with some probabilities ties, we relax the
ranking comparison with the decision function.
Non-regression test for:
https://github.com/scikit-learn/scikit-learn/issues/24025
"""
estimator = SGDClassifier(loss="log_loss")
check_decision_proba_consistency("SGDClassifier", estimator)