Skip to content

FIX ElasticNet.fit does not modify sample_weight in place #19055

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 6 commits into from
Dec 22, 2020
Merged

FIX ElasticNet.fit does not modify sample_weight in place #19055

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
30171f6
BUG ElasticNet.fit does not modify sample_weight in place
thomasjpfan Dec 21, 2020
bdd14ba
DOC Adds whats new
thomasjpfan Dec 21, 2020
7868c59
DOC Adds name to whats new
thomasjpfan Dec 22, 2020
d486a65
DOC Fixes spelling mistakes
thomasjpfan Dec 22, 2020
8c192ae
TST Be extra cautious
thomasjpfan Dec 22, 2020
bfa4a1a
Merge remote-tracking branch 'upstream/master' into elastic_net_sampl…
thomasjpfan Dec 22, 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
6 changes: 6 additions & 0 deletions doc/whats_new/v1.0.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,6 +56,12 @@ Changelog
in multicore settings. :pr:`19052` by
:user:`Yusuke Nagasaka <YusukeNagasaka>`.

:mod:`sklearn.linear_model`
...........................

- |Fix| :meth:`ElasticNet.fit` no longer modifies `sample_weight` in place.
:pr:`19055` by `Thomas Fan`_.

Code and Documentation Contributors
-----------------------------------

Expand Down
2 changes: 1 addition & 1 deletion sklearn/linear_model/_coordinate_descent.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -790,7 +790,7 @@ def fit(self, X, y, sample_weight=None, check_input=True):
dtype=X.dtype)
# simplify things by rescaling sw to sum up to n_samples
# => np.average(x, weights=sw) = np.mean(sw * x)
sample_weight *= (n_samples / np.sum(sample_weight))
sample_weight = sample_weight * (n_samples / np.sum(sample_weight))
# Objective function is:
# 1/2 * np.average(squared error, weights=sw) + alpha * penalty
# but coordinate descent minimizes:
Expand Down
19 changes: 19 additions & 0 deletions sklearn/linear_model/tests/test_coordinate_descent.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1214,3 +1214,22 @@ def test_linear_models_cv_fit_for_all_backends(backend, estimator):

with joblib.parallel_backend(backend=backend):
estimator(n_jobs=2, cv=3).fit(X, y)


@pytest.mark.parametrize("check_input", [True, False])
def test_enet_sample_weight_does_not_overwrite_sample_weight(check_input):
"""Check that ElasticNet does not overwrite sample_weights."""

rng = np.random.RandomState(0)
n_samples, n_features = 10, 5

X = rng.rand(n_samples, n_features)
y = rng.rand(n_samples)

sample_weight_1_25 = 1.25 * np.ones_like(y)
sample_weight = sample_weight_1_25.copy()

reg = ElasticNet()
reg.fit(X, y, sample_weight=sample_weight, check_input=check_input)

assert_array_equal(sample_weight, sample_weight_1_25)