[go: nahoru, domu]
Skip to content
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

Jump to bottom

ENH Consistent loss name for absolute error #19733

Merged
merged 13 commits into from
May 10, 2021
Merged

ENH Consistent loss name for absolute error #19733

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
b22aa54
MNT deprecate mae criterion in tree module
lorentzenchr Mar 20, 2021
0bdeb7b
MNT deprecate loss absolute_loss in RANSACRegressor
lorentzenchr Mar 20, 2021
4fe9982
MNT deprecate loss least_absolute_deviation in HGBRegressor
lorentzenchr Mar 20, 2021
b9f5677
MNT deprecate criterion mae in forests
lorentzenchr Mar 20, 2021
9c525e0
MNT deprecate loss lad in GradientBoostingRegressor
lorentzenchr Mar 20, 2021
a6b2796
DOC replace least_absolute_deviation in user guide
lorentzenchr Mar 20, 2021
3254391
DOC add whatsnew entry
lorentzenchr Mar 20, 2021
2e65634
TST nicer deprecation tests
lorentzenchr Mar 20, 2021
f6f1816
MNT deal with mae in _make_estimator
lorentzenchr Mar 20, 2021
9a8d1c7
TST replace remaining instances of "mae" in test_forest.py
lorentzenchr Mar 21, 2021
125162d
TST nicer depcrecation test for HGBT
lorentzenchr Mar 21, 2021
707453e
Merge branch 'main' into consistent_absolute_error
lorentzenchr Apr 22, 2021
1aeed2b
Merge branch 'main' into consistent_absolute_error
lorentzenchr May 1, 2021
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
Prev Previous commit
Next Next commit
TST nicer deprecation tests
  • Loading branch information
@lorentzenchr
lorentzenchr committed Mar 20, 2021
commit 2e6563488ffba14332585b342a748973e230bc6f
14 changes: 8 additions & 6 deletions sklearn/ensemble/tests/test_forest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1498,16 +1498,18 @@ def test_n_features_deprecation(Estimator):


# TODO: Remove in v1.2
@pytest.mark.parametrize("criterion", ["mse", "mae"])
def test_criterion_deprecated(criterion):
est1 = RandomForestRegressor(criterion=criterion, random_state=0)
@pytest.mark.parametrize("old_criterion, new_criterion", [
("mse", "squared_error"),
("mae", "absolute_error"),
])
def test_criterion_deprecated(old_criterion, new_criterion):
est1 = RandomForestRegressor(criterion=old_criterion, random_state=0)

with pytest.warns(FutureWarning,
match="Criterion '" + criterion + "' was deprecated"):
match=f"Criterion '{old_criterion}' was deprecated"):
est1.fit(X, y)

d = {"mse": "squared_error", "mae": "absolute_error"}
est2 = RandomForestRegressor(criterion=d[criterion], random_state=0)
est2 = RandomForestRegressor(criterion=new_criterion, random_state=0)
est2.fit(X, y)
assert_allclose(est1.predict(X), est2.predict(X))

Expand Down
14 changes: 8 additions & 6 deletions sklearn/ensemble/tests/test_gradient_boosting.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1390,15 +1390,17 @@ def test_criterion_mse_deprecated(Estimator):


# TODO: Remove in v1.2
@pytest.mark.parametrize("loss", ["ls", "lad"])
def test_loss_deprecated(loss):
est1 = GradientBoostingRegressor(loss=loss, random_state=0)
@pytest.mark.parametrize("old_loss, new_loss", [
("ls", "squared_error"),
("lad", "absolute_error"),
])
def test_loss_deprecated(old_loss, new_loss):
est1 = GradientBoostingRegressor(loss=old_loss, random_state=0)

with pytest.warns(FutureWarning,
match="The loss '" + loss + "' was deprecated"):
match=f"The loss '{old_loss}' was deprecated"):
est1.fit(X, y)

d = {"ls": "squared_error", "lad": "absolute_error"}
est2 = GradientBoostingRegressor(loss=d[loss], random_state=0)
est2 = GradientBoostingRegressor(loss=new_loss, random_state=0)
est2.fit(X, y)
assert_allclose(est1.predict(X), est2.predict(X))
14 changes: 8 additions & 6 deletions sklearn/linear_model/tests/test_ransac.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -539,15 +539,17 @@ def test_ransac_final_model_fit_sample_weight():


# TODO: Remove in v1.2
@pytest.mark.parametrize("loss", ["absolute_loss", "squared_loss"])
def test_loss_deprecated(loss):
est1 = RANSACRegressor(loss=loss, random_state=0)
@pytest.mark.parametrize("old_loss, new_loss", [
("absolute_loss", "squared_error"),
("squared_loss", "absolute_error"),
])
def test_loss_deprecated(old_loss, new_loss):
est1 = RANSACRegressor(loss=old_loss, random_state=0)

with pytest.warns(FutureWarning,
match="The loss '" + loss + "' was deprecated"):
match=f"The loss '{old_loss}' was deprecated"):
est1.fit(X, y)

d = {"absolute_loss": "absolute_error", "squared_loss": "squared_error"}
est2 = RANSACRegressor(loss=d[loss], random_state=0)
est2 = RANSACRegressor(loss=new_loss, random_state=0)
est2.fit(X, y)
assert_allclose(est1.predict(X), est2.predict(X))
14 changes: 8 additions & 6 deletions sklearn/tree/tests/test_tree.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2121,14 +2121,16 @@ def test_X_idx_sorted_deprecated(TreeEstimator):

# TODO: Remove in v1.2
@pytest.mark.parametrize("Tree", REG_TREES.values())
@pytest.mark.parametrize("criterion", ["mse", "mae"])
def test_mse_deprecated(Tree, criterion):
tree = Tree(criterion=criterion)
@pytest.mark.parametrize("old_criterion, new_criterion", [
("mse", "squared_error"),
("mae", "absolute_error"),
])
def test_criterion_deprecated(Tree, old_criterion, new_criterion):
tree = Tree(criterion=old_criterion)

with pytest.warns(FutureWarning,
match="Criterion '" + criterion + "' was deprecated"):
match=f"Criterion '{old_criterion}' was deprecated"):
tree.fit(X, y)

d = {"mse": "squared_error", "mae": "absolute_error"}
tree_new = Tree(criterion=d[criterion]).fit(X, y)
tree_new = Tree(criterion=new_criterion).fit(X, y)
assert_allclose(tree.predict(X), tree_new.predict(X))