iter

Author: glemaitre

README.rst

@@ -66,6 +66,7 @@ Dependencies
 - NumPy (>= |NumPyMinVersion|)
 - SciPy (>= |SciPyMinVersion|)
 - Scikit-learn (>= |ScikitLearnMinVersion|)
+- Pytest (>= |PytestMinVersion|)
 
 Additionally, `imbalanced-learn` requires the following optional dependencies:
 

imblearn/_min_dependencies.py

@@ -6,7 +6,6 @@
 
 import copy
 import numbers
-import warnings
 
 import numpy as np
 import sklearn
@@ -121,14 +120,6 @@ class BalancedBaggingClassifier(_ParamsValidationMixin, BaggingClassifier):
 
         .. versionadded:: 0.10
 
-    n_features_ : int
-        The number of features when `fit` is performed.
-
-        .. deprecated:: 1.0
-           `n_features_` is deprecated in `scikit-learn` 1.0 and will be removed
-           in version 1.2. When the minimum version of `scikit-learn` supported
-           by `imbalanced-learn` will reach 1.2, this attribute will be removed.
-
     estimators_ : list of estimators
         The collection of fitted base estimators.
 
@@ -331,20 +322,6 @@ def _validate_estimator(self, default=DecisionTreeClassifier()):
             [("sampler", self.sampler_), ("classifier", estimator)]
         )
 
-    # TODO: remove when supporting scikit-learn>=1.2
-    @property
-    def n_features_(self):
-        """Number of features when ``fit`` is performed."""
-        warnings.warn(
-            (
-                "`n_features_` was deprecated in scikit-learn 1.0. This attribute will "
-                "not be accessible when the minimum supported version of scikit-learn "
-                "is 1.2."
-            ),
-            FutureWarning,
-        )
-        return self.n_features_in_
-
     @_fit_context(prefer_skip_nested_validation=False)
     def fit(self, X, y):
         """Build a Bagging ensemble of estimators from the training set (X, y).

imblearn/ensemble/_bagging.py

@@ -100,14 +100,6 @@ class EasyEnsembleClassifier(_ParamsValidationMixin, BaggingClassifier):
     n_classes_ : int or list
         The number of classes.
 
-    n_features_ : int
-        The number of features when `fit` is performed.
-
-        .. deprecated:: 1.0
-           `n_features_` is deprecated in `scikit-learn` 1.0 and will be removed
-           in version 1.2. When the minimum version of `scikit-learn` supported
-           by `imbalanced-learn` will reach 1.2, this attribute will be removed.
-
     n_features_in_ : int
         Number of features in the input dataset.
 

imblearn/ensemble/_easy_ensemble.py

@@ -343,14 +343,6 @@ class labels (multi-output problem).
         The number of classes (single output problem), or a list containing the
         number of classes for each output (multi-output problem).
 
-    n_features_ : int
-        The number of features when `fit` is performed.
-
-        .. deprecated:: 1.0
-           `n_features_` is deprecated in `scikit-learn` 1.0 and will be removed
-           in version 1.2. When the minimum version of `scikit-learn` supported
-           by `imbalanced-learn` will reach 1.2, this attribute will be removed.
-
     n_features_in_ : int
         Number of features in the input dataset.
 
@@ -502,13 +494,8 @@ def __init__(
     def _validate_estimator(self, default=DecisionTreeClassifier()):
         """Check the estimator and the n_estimator attribute, set the
         `estimator_` attribute."""
-        if hasattr(self, "estimator"):
-            base_estimator = self.estimator
-        else:
-            base_estimator = self.base_estimator
-
-        if base_estimator is not None:
-            self.estimator_ = clone(base_estimator)
+        if self.estimator is not None:
+            self.estimator_ = clone(self.estimator)
         else:
             self.estimator_ = clone(default)
 
@@ -893,22 +880,5 @@ def _compute_oob_predictions(self, X, y):
 
         return oob_pred
 
-    # TODO: remove when supporting scikit-learn>=1.2
-    @property
-    def n_features_(self):
-        """Number of features when ``fit`` is performed."""
-        warn(
-            (
-                "`n_features_` was deprecated in scikit-learn 1.0. This attribute will "
-                "not be accessible when the minimum supported version of scikit-learn "
-                "is 1.2."
-            ),
-            FutureWarning,
-        )
-        return self.n_features_in_
-
     def _more_tags(self):
-        return {
-            "multioutput": False,
-            "multilabel": False,
-        }
+        return {"multioutput": False, "multilabel": False}

imblearn/ensemble/_forest.py

@@ -581,11 +581,3 @@ def roughly_balanced_bagging(X, y, replace=False):
     for estimator in rbb.estimators_:
         class_counts = estimator[-1].class_counts_
         assert (class_counts[0] / class_counts[1]) > 0.78
-
-
-def test_balanced_bagging_classifier_n_features():
-    """Check that we raise a FutureWarning when accessing `n_features_`."""
-    X, y = load_iris(return_X_y=True)
-    estimator = BalancedBaggingClassifier().fit(X, y)
-    with pytest.warns(FutureWarning, match="`n_features_` was deprecated"):
-        estimator.n_features_

imblearn/ensemble/tests/test_bagging.py

@@ -222,11 +222,3 @@ def test_easy_ensemble_classifier_grid_search():
         cv=5,
     )
     grid_search.fit(X, y)
-
-
-def test_easy_ensemble_classifier_n_features():
-    """Check that we raise a FutureWarning when accessing `n_features_`."""
-    X, y = load_iris(return_X_y=True)
-    estimator = EasyEnsembleClassifier().fit(X, y)
-    with pytest.warns(FutureWarning, match="`n_features_` was deprecated"):
-        estimator.n_features_

imblearn/ensemble/tests/test_easy_ensemble.py

@@ -1,7 +1,7 @@
 import numpy as np
 import pytest
 import sklearn
-from sklearn.datasets import load_iris, make_classification
+from sklearn.datasets import make_classification
 from sklearn.model_selection import GridSearchCV, train_test_split
 from sklearn.utils._testing import assert_allclose, assert_array_equal
 from sklearn.utils.fixes import parse_version
@@ -219,16 +219,6 @@ def test_balanced_random_forest_oob_binomial(ratio):
     assert np.abs(erf.oob_score_ - 0.5) < 0.1
 
 
-def test_balanced_bagging_classifier_n_features():
-    """Check that we raise a FutureWarning when accessing `n_features_`."""
-    X, y = load_iris(return_X_y=True)
-    estimator = BalancedRandomForestClassifier(
-        sampling_strategy="all", replacement=True, bootstrap=False
-    ).fit(X, y)
-    with pytest.warns(FutureWarning, match="`n_features_` was deprecated"):
-        estimator.n_features_
-
-
 # TODO: remove in 0.13
 def test_balanced_random_forest_change_behaviour(imbalanced_dataset):
     """Check that we raise a change of behaviour for the parameters `sampling_strategy`

imblearn/ensemble/tests/test_forest.py

@@ -73,10 +73,7 @@ def _set_checking_parameters(estimator):
     if "n_estimators" in params:
         estimator.set_params(n_estimators=min(5, estimator.n_estimators))
     if name == "ClusterCentroids":
-        if sklearn_version < parse_version("1.1"):
-            algorithm = "full"
-        else:
-            algorithm = "lloyd"
+        algorithm = "lloyd"
         estimator.set_params(
             voting="soft",
             estimator=KMeans(random_state=0, algorithm=algorithm, n_init=1),
@@ -689,33 +686,14 @@ def check_dataframe_column_names_consistency(name, estimator_orig):
         X_bad = pd.DataFrame(X, columns=invalid_name)
 
         for name, method in check_methods:
-            if sklearn_version >= parse_version("1.2"):
-                expected_msg = re.escape(
-                    "The feature names should match those that were passed during fit."
-                    f"\n{additional_message}"
-                )
-                with raises(
-                    ValueError, match=expected_msg, err_msg=f"{name} did not raise"
-                ):
-                    method(X_bad)
-            else:
-                expected_msg = re.escape(
-                    "The feature names should match those that were passed "
-                    "during fit. Starting version 1.2, an error will be raised.\n"
-                    f"{additional_message}"
-                )
-                with warnings.catch_warnings():
-                    warnings.filterwarnings(
-                        "error",
-                        category=FutureWarning,
-                        module="sklearn",
-                    )
-                    with raises(
-                        FutureWarning,
-                        match=expected_msg,
-                        err_msg=f"{name} did not raise",
-                    ):
-                        method(X_bad)
+            expected_msg = re.escape(
+                "The feature names should match those that were passed during fit."
+                f"\n{additional_message}"
+            )
+            with raises(
+                ValueError, match=expected_msg, err_msg=f"{name} did not raise"
+            ):
+                method(X_bad)
 
         # partial_fit checks on second call
         # Do not call partial fit if early_stopping is on

imblearn/utils/estimator_checks.py

@@ -1,14 +1,16 @@
 """Tests for the minimum dependencies in the README.rst file."""
+
 import os
 import platform
 import re
 from pathlib import Path
 
 import pytest
-from sklearn.utils.fixes import parse_version
+import tomllib
+from packaging.requirements import Requirement
+from packaging.version import parse
 
 import imblearn
-from imblearn._min_dependencies import dependent_packages
 
 
 @pytest.mark.skipif(
@@ -17,7 +19,35 @@
 def test_min_dependencies_readme():
     # Test that the minimum dependencies in the README.rst file are
     # consistent with the minimum dependencies defined at the file:
-    # imblearn/_min_dependencies.py
+    # pyproject.toml
+
+    pyproject_path = Path(imblearn.__path__[0]).parents[0] / "pyproject.toml"
+    with open(pyproject_path, "rb") as f:
+        pyproject_data = tomllib.load(f)
+
+    def process_requirements(requirements):
+        result = {}
+        for req in requirements:
+            req = Requirement(req)
+            for specifier in req.specifier:
+                if specifier.operator == ">=":
+                    result[req.name] = parse(specifier.version)
+        return result
+
+    min_dependencies = process_requirements(
+        [f"python{pyproject_data['project']['requires-python']}"]
+    )
+    min_dependencies.update(
+        process_requirements(pyproject_data["project"]["dependencies"])
+    )
+
+    markers = ["docs", "optional", "tensorflow", "keras", "tests"]
+    for marker_name in markers:
+        min_dependencies.update(
+            process_requirements(
+                pyproject_data["project"]["optional-dependencies"][marker_name]
+            )
+        )
 
     pattern = re.compile(
         r"(\.\. \|)"
@@ -44,8 +74,8 @@ def test_min_dependencies_readme():
             package, version = matched.group(2), matched.group(5)
             package = package.lower()
 
-            if package in dependent_packages:
-                version = parse_version(version)
-                min_version = parse_version(dependent_packages[package][0])
+            if package in min_dependencies:
+                version = parse(version)
+                min_version = min_dependencies[package]
 
                 assert version == min_version, f"{package} has a mismatched version"