Skip to content

Deprecate use of probability=True in SVC and NuSVC #32050

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

Open
wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Deprecate use of probability=True in SVC and NuSVC #32050

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
d5cc859
lazy test added
snath-xoc Feb 28, 2025
904f0fe
merge main
snath-xoc Jun 13, 2025
08da1e2
initial commit
snath-xoc Aug 29, 2025
b3c5314
Merge branch 'main' into deprecate_porb_NuSVC
snath-xoc Aug 29, 2025
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
30 changes: 29 additions & 1 deletion sklearn/svm/_classes.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,7 @@
# Authors: The scikit-learn developers
# SPDX-License-Identifier: BSD-3-Clause

import warnings
from numbers import Integral, Real

import numpy as np
Expand Down Expand Up @@ -690,6 +691,10 @@ class SVC(BaseSVC):
5-fold cross-validation, and `predict_proba` may be inconsistent with
`predict`. Read more in the :ref:`User Guide <scores_probabilities>`.

..deprecated:: version 1.8 parameter probability will be deprecated in future
versions, use CalibratedClassifierCV(SVC(), ensemble=False) instead
of SVC(probability=True)

tol : float, default=1e-3
Tolerance for stopping criterion.

Expand Down Expand Up @@ -867,7 +872,7 @@ def __init__(
gamma="scale",
coef0=0.0,
shrinking=True,
probability=False,
probability="deprecated",
tol=1e-3,
cache_size=200,
class_weight=None,
Expand All @@ -877,6 +882,16 @@ def __init__(
break_ties=False,
random_state=None,
):
if probability != "deprecated":
warnings.warn(
"parameter `probability` will be deprecated in version 1.8, "
"use `CalibratedClassifierCV(SVC(), ensemble=False)` instead "
"of `SVC(probability=True)`",
FutureWarning,
)
else:
probability = False

super().__init__(
kernel=kernel,
degree=degree,
Expand Down Expand Up @@ -951,6 +966,10 @@ class NuSVC(BaseSVC):
5-fold cross-validation, and `predict_proba` may be inconsistent with
`predict`. Read more in the :ref:`User Guide <scores_probabilities>`.

..deprecated:: version 1.8 parameter probability will be deprecated in future
versions, use CalibratedClassifierCV(NuSVC(), ensemble=False) instead
of NuSVC(probability=True)

tol : float, default=1e-3
Tolerance for stopping criterion.

Expand Down Expand Up @@ -1140,6 +1159,15 @@ def __init__(
break_ties=False,
random_state=None,
):
if probability != "deprecated":
warnings.warn(
"parameter `probability` will be deprecated in version 1.8, "
"use `CalibratedClassifierCV(NuSVC(), ensemble=False)` instead "
"of `NuSVC(probability=True)`",
FutureWarning,
)
else:
probability = False
super().__init__(
kernel=kernel,
degree=degree,
Expand Down
16 changes: 16 additions & 0 deletions sklearn/svm/tests/test_svm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -379,6 +379,7 @@ def test_tweak_params():
# XXX: this test is thread-unsafe because it uses probability=True:
# https://github.com/scikit-learn/scikit-learn/issues/31885
@pytest.mark.thread_unsafe
@pytest.mark.filterwarnings("ignore::FutureWarning")
def test_probability(global_random_seed):
# Predict probabilities using SVC
# This uses cross validation, so we use a slightly bigger testing set.
Expand Down Expand Up @@ -764,6 +765,7 @@ def test_svc_nonfinite_params(global_random_seed):
clf.fit(X, y)


@pytest.mark.filterwarnings("ignore::FutureWarning")
def test_unicode_kernel(global_random_seed):
# Test that a unicode kernel name does not cause a TypeError
iris = get_iris_dataset(global_random_seed)
Expand Down Expand Up @@ -1055,6 +1057,7 @@ def test_linearsvc_verbose():
# XXX: this test is thread-unsafe because it uses probability=True:
# https://github.com/scikit-learn/scikit-learn/issues/31885
@pytest.mark.thread_unsafe
@pytest.mark.filterwarnings("ignore::FutureWarning")
def test_svc_clone_with_callable_kernel():
iris = get_iris_dataset(42)

Expand Down Expand Up @@ -1102,6 +1105,7 @@ def test_svc_bad_kernel():
# XXX: this test is thread-unsafe because it uses probability=True:
# https://github.com/scikit-learn/scikit-learn/issues/31885
@pytest.mark.thread_unsafe
@pytest.mark.filterwarnings("ignore::FutureWarning")
def test_libsvm_convergence_warnings(global_random_seed):
a = svm.SVC(
kernel=lambda x, y: np.dot(x, y.T),
Expand Down Expand Up @@ -1135,6 +1139,7 @@ def test_unfitted():
# https://github.com/scikit-learn/scikit-learn/issues/31885
@pytest.mark.thread_unsafe
@pytest.mark.filterwarnings("ignore::sklearn.exceptions.ConvergenceWarning")
@pytest.mark.filterwarnings("ignore::FutureWarning")
def test_consistent_proba(global_random_seed):
a = svm.SVC(probability=True, max_iter=1, random_state=global_random_seed)
proba_1 = a.fit(X, Y).predict_proba(X)
Expand Down Expand Up @@ -1190,6 +1195,7 @@ def test_lsvc_intercept_scaling_zero():
assert lsvc.intercept_ == 0.0


@pytest.mark.filterwarnings("ignore::FutureWarning")
def test_hasattr_predict_proba(global_random_seed):
iris = get_iris_dataset(global_random_seed)

Expand Down Expand Up @@ -1534,3 +1540,13 @@ def test_svm_with_infinite_C(Estimator, make_dataset, C_inf, global_random_seed)
estimator_C_large = Estimator(C=1e10).fit(X, y)

assert_allclose(estimator_C_large.predict(X), estimator_C_inf.predict(X))


@pytest.mark.parametrize(
"Estimator, name",
[(svm.SVC, "SVC"), (svm.NuSVC, "NuSVC")],
)
def test_probability_raises_futurewarning(Estimator, name):
X, y = make_classification()
with pytest.warns(FutureWarning):
Estimator(probability=True).fit(X, y)
Loading