Skip to content

ENH Remove unnecessary OOB computation when n_more_estimators == 0 #26318

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 11 commits into from
May 30, 2023
Merged

ENH Remove unnecessary OOB computation when n_more_estimators == 0 #26318

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
e0421ec
Edited docs for warm_start and removed OOB computation for n_estimato…
Oct 4, 2022
d6d7832
Removed change to OOB computation
Nov 12, 2022
aff6597
Removed OOB computation when n_more_estimators == 0
May 2, 2023
7e8a923
Added test to check for unnecessary OOB computation when n_more_estim…
May 23, 2023
3d5a712
Removed change to OOB computation
Nov 12, 2022
5e74e3a
Removed OOB computation when n_more_estimators == 0
May 2, 2023
b3c6c7e
Split tests and updated changelog
May 27, 2023
3ef0393
Corrected with linter
May 27, 2023
3b1c708
Updated changelog and refactored unit test
May 29, 2023
2e05fb7
Fixed accidental deletion of old unit tests
May 29, 2023
75cab60
Apply suggestions from code review
glemaitre May 30, 2023
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/v1.3.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -287,6 +287,12 @@ Changelog
scikit-learn 1.3: retraining with scikit-learn 1.3 is required.
:pr:`25186` by :user:`Felipe Breve Siola <fsiola>`.

- |Efficiency| :class:`ensemble.RandomForestClassifier` and
:class:`ensemble.RandomForestRegressor` with `warm_start=True` now only
recomputes out-of-bag scores when there are actually more `n_estimators`
in subsequent `fit` calls.
:pr:`26318` by :user:`Joshua Choo Yun Keat <choo8>`.

- |Enhancement| :class:`ensemble.BaggingClassifier` and
:class:`ensemble.BaggingRegressor` expose the `allow_nan` tag from the
underlying estimator. :pr:`25506` by `Thomas Fan`_.
Expand Down
4 changes: 3 additions & 1 deletion sklearn/ensemble/_forest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -474,7 +474,9 @@ def fit(self, X, y, sample_weight=None):
# Collect newly grown trees
self.estimators_.extend(trees)

if self.oob_score:
if self.oob_score and (
n_more_estimators > 0 or not hasattr(self, "oob_score_")
):
y_type = type_of_target(y)
if y_type in ("multiclass-multioutput", "unknown"):
# FIXME: we could consider to support multiclass-multioutput if
Expand Down
22 changes: 22 additions & 0 deletions sklearn/ensemble/tests/test_forest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -58,6 +58,7 @@

from sklearn.tree._classes import SPARSE_SPLITTERS

from unittest.mock import patch

# toy sample
X = [[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1]]
Expand Down Expand Up @@ -1470,6 +1471,27 @@ def test_warm_start_oob(name):
check_warm_start_oob(name)


@pytest.mark.parametrize("name", FOREST_CLASSIFIERS_REGRESSORS)
def test_oob_not_computed_twice(name):
# Check that oob_score is not computed twice when warm_start=True.
X, y = hastie_X, hastie_y
ForestEstimator = FOREST_ESTIMATORS[name]

est = ForestEstimator(
n_estimators=10, warm_start=True, bootstrap=True, oob_score=True
)

with patch.object(
est, "_set_oob_score_and_attributes", wraps=est._set_oob_score_and_attributes
) as mock_set_oob_score_and_attributes:
est.fit(X, y)

with pytest.warns(UserWarning, match="Warm-start fitting without increasing"):
est.fit(X, y)

mock_set_oob_score_and_attributes.assert_called_once()


def test_dtype_convert(n_classes=15):
classifier = RandomForestClassifier(random_state=0, bootstrap=False)

Expand Down