MAINT prepare release 0.14 (#1146)

Author: glemaitre

README.rst

@@ -28,13 +28,13 @@
 .. _Black: :target: https://github.com/psf/black
 
 .. |PythonMinVersion| replace:: 3.10
-.. |NumPyMinVersion| replace:: 1.24.3
-.. |SciPyMinVersion| replace:: 1.10.1
-.. |ScikitLearnMinVersion| replace:: 1.3.2
+.. |NumPyMinVersion| replace:: 1.25.2
+.. |SciPyMinVersion| replace:: 1.11.4
+.. |ScikitLearnMinVersion| replace:: 1.4.2
 .. |MatplotlibMinVersion| replace:: 3.7.3
-.. |PandasMinVersion| replace:: 1.5.3
-.. |TensorflowMinVersion| replace:: 2.13.1
-.. |KerasMinVersion| replace:: 3.0.5
+.. |PandasMinVersion| replace:: 2.0.3
+.. |TensorflowMinVersion| replace:: 2.16.1
+.. |KerasMinVersion| replace:: 3.3.3
 .. |SeabornMinVersion| replace:: 0.12.2
 .. |PytestMinVersion| replace:: 7.2.2
 

doc/install.rst

@@ -8,13 +8,13 @@ Prerequisites
 =============
 
 .. |PythonMinVersion| replace:: 3.10
-.. |NumPyMinVersion| replace:: 1.24.3
-.. |SciPyMinVersion| replace:: 1.10.1
-.. |ScikitLearnMinVersion| replace:: 1.3.2
+.. |NumPyMinVersion| replace:: 1.25.2
+.. |SciPyMinVersion| replace:: 1.11.4
+.. |ScikitLearnMinVersion| replace:: 1.4.2
 .. |MatplotlibMinVersion| replace:: 3.7.3
-.. |PandasMinVersion| replace:: 1.5.3
-.. |TensorflowMinVersion| replace:: 2.13.1
-.. |KerasMinVersion| replace:: 3.0.5
+.. |PandasMinVersion| replace:: 2.0.3
+.. |TensorflowMinVersion| replace:: 2.16.1
+.. |KerasMinVersion| replace:: 3.3.3
 .. |SeabornMinVersion| replace:: 0.12.2
 .. |PytestMinVersion| replace:: 7.2.2
 

doc/whats_new/0.14.rst

@@ -1,9 +1,9 @@
 .. _changes_0_14:
 
-Version 0.14.0 (Under development)
-==================================
+Version 0.14.0
+==============
 
-**TBD**
+**August 14, 2025**
 
 Changelog
 ---------
@@ -21,5 +21,8 @@ Enhancements
 Compatibility
 .............
 
+- Compatibility with scikit-learn 1.7
+  :pr:`1137`, :pr:`1145`, :pr:`1146` by :user:`Guillaume Lemaitre <glemaitre>`.
+
 Deprecations
 ............

examples/model_selection/plot_instance_hardness_cv.py

@@ -28,15 +28,15 @@
 from sklearn.datasets import make_blobs
 
 X, y = make_blobs(n_samples=[950, 50], centers=((-3, 0), (3, 0)), random_state=10)
-plt.scatter(X[:, 0], X[:, 1], c=y)
+_ = plt.scatter(X[:, 0], X[:, 1], c=y)
 
 # %%
 # To introduce instance hardness in our dataset, we add some hard to classify samples:
 X_hard, y_hard = make_blobs(
     n_samples=10, centers=((3, 0), (-3, 0)), cluster_std=1, random_state=10
 )
 X, y = np.vstack((X, X_hard)), np.hstack((y, y_hard))
-plt.scatter(X[:, 0], X[:, 1], c=y)
+_ = plt.scatter(X[:, 0], X[:, 1], c=y)
 
 # %%
 # Compare cross validation scores using `StratifiedKFold` and `InstanceHardnessCV`
@@ -69,7 +69,7 @@
 results = {}
 for cv in (
     StratifiedKFold(n_splits=5, shuffle=True, random_state=10),
-    InstanceHardnessCV(estimator=LogisticRegression(), n_splits=5, random_state=10),
+    InstanceHardnessCV(estimator=LogisticRegression()),
 ):
     result = cross_validate(
         logistic_regression,
@@ -83,7 +83,7 @@
 
 # %%
 ax = results.plot.box(vert=False, whis=[0, 100])
-ax.set(
+_ = ax.set(
     xlabel="Average precision",
     title="Cross validation scores with different splitters",
     xlim=(0, 1),

imblearn/ensemble/_bagging.py

@@ -12,15 +12,13 @@
 from sklearn.ensemble import BaggingClassifier
 from sklearn.tree import DecisionTreeClassifier
 from sklearn.utils._param_validation import HasMethods, Interval, StrOptions
-from sklearn.utils.fixes import parse_version
 
 from ..pipeline import Pipeline
 from ..under_sampling import RandomUnderSampler
 from ..under_sampling.base import BaseUnderSampler
 from ..utils import Substitution, check_sampling_strategy, check_target_type
 from ..utils._docstring import _n_jobs_docstring, _random_state_docstring
-from ..utils._sklearn_compat import _fit_context, sklearn_version
-from ._common import _bagging_parameter_constraints
+from ..utils._sklearn_compat import _fit_context
 
 
 @Substitution(
@@ -224,11 +222,7 @@ class BalancedBaggingClassifier(BaggingClassifier):
     """
 
     # make a deepcopy to not modify the original dictionary
-    if sklearn_version >= parse_version("1.4"):
-        _parameter_constraints = copy.deepcopy(BaggingClassifier._parameter_constraints)
-    else:
-        _parameter_constraints = copy.deepcopy(_bagging_parameter_constraints)
-
+    _parameter_constraints = copy.deepcopy(BaggingClassifier._parameter_constraints)
     _parameter_constraints.update(
         {
             "sampling_strategy": [
@@ -241,9 +235,6 @@ class BalancedBaggingClassifier(BaggingClassifier):
             "sampler": [HasMethods(["fit_resample"]), None],
         }
     )
-    # TODO: remove when minimum supported version of scikit-learn is 1.4
-    if "base_estimator" in _parameter_constraints:
-        del _parameter_constraints["base_estimator"]
 
     def __init__(
         self,

imblearn/ensemble/_common.py

@@ -19,10 +19,8 @@ def _estimator_has(attr):
     def check(self):
         if hasattr(self, "estimators_"):
             return hasattr(self.estimators_[0], attr)
-        elif self.estimator is not None:
+        else:  # self.estimator is not None
             return hasattr(self.estimator, attr)
-        else:  # TODO(1.4): Remove when the base_estimator deprecation cycle ends
-            return hasattr(self.base_estimator, attr)
 
     return check
 
@@ -45,11 +43,6 @@ def check(self):
     "n_jobs": [None, Integral],
     "random_state": ["random_state"],
     "verbose": ["verbose"],
-    "base_estimator": [
-        HasMethods(["fit", "predict"]),
-        StrOptions({"deprecated"}),
-        None,
-    ],
 }
 
 _adaboost_classifier_parameter_constraints = {

imblearn/ensemble/_easy_ensemble.py

@@ -5,19 +5,13 @@
 # License: MIT
 
 import copy
-import inspect
 import numbers
 
 import numpy as np
 from sklearn.base import clone
 from sklearn.ensemble import AdaBoostClassifier, BaggingClassifier
-from sklearn.ensemble._bagging import _parallel_decision_function
-from sklearn.ensemble._base import _partition_estimators
 from sklearn.utils._param_validation import Interval, StrOptions
 from sklearn.utils.fixes import parse_version
-from sklearn.utils.metaestimators import available_if
-from sklearn.utils.parallel import Parallel, delayed
-from sklearn.utils.validation import check_is_fitted
 
 from ..pipeline import Pipeline
 from ..under_sampling import RandomUnderSampler
@@ -28,9 +22,8 @@
     _fit_context,
     get_tags,
     sklearn_version,
-    validate_data,
 )
-from ._common import _bagging_parameter_constraints, _estimator_has
+from ._common import _bagging_parameter_constraints
 
 MAX_INT = np.iinfo(np.int32).max
 
@@ -276,61 +269,6 @@ def _fit(self, X, y, max_samples=None, max_depth=None, sample_weight=None):
         # None.
         return super()._fit(X, y, self.max_samples)
 
-    # TODO: remove when minimum supported version of scikit-learn is 1.1
-    @available_if(_estimator_has("decision_function"))
-    def decision_function(self, X):
-        """Average of the decision functions of the base classifiers.
-
-        Parameters
-        ----------
-        X : {array-like, sparse matrix} of shape (n_samples, n_features)
-            The training input samples. Sparse matrices are accepted only if
-            they are supported by the base estimator.
-
-        Returns
-        -------
-        score : ndarray of shape (n_samples, k)
-            The decision function of the input samples. The columns correspond
-            to the classes in sorted order, as they appear in the attribute
-            ``classes_``. Regression and binary classification are special
-            cases with ``k == 1``, otherwise ``k==n_classes``.
-        """
-        check_is_fitted(self)
-
-        # Check data
-        X = validate_data(
-            self,
-            X=X,
-            accept_sparse=["csr", "csc"],
-            dtype=None,
-            ensure_all_finite=(
-                "allow_nan" if get_tags(self).input_tags.allow_nan else True
-            ),
-            reset=False,
-        )
-
-        # Parallel loop
-        n_jobs, _, starts = _partition_estimators(self.n_estimators, self.n_jobs)
-
-        kwargs = {}
-        if "params" in inspect.signature(_parallel_decision_function).parameters:
-            kwargs["params"] = {}
-
-        all_decisions = Parallel(n_jobs=n_jobs, verbose=self.verbose)(
-            delayed(_parallel_decision_function)(
-                self.estimators_[starts[i] : starts[i + 1]],
-                self.estimators_features_[starts[i] : starts[i + 1]],
-                X,
-                **kwargs,
-            )
-            for i in range(n_jobs)
-        )
-
-        # Reduce
-        decisions = sum(all_decisions) / self.n_estimators
-
-        return decisions
-
     @property
     def base_estimator_(self):
         """Attribute for older sklearn version compatibility."""

imblearn/ensemble/_forest.py

@@ -74,12 +74,9 @@ def _local_parallel_build_trees(
         "bootstrap": bootstrap,
     }
 
-    if sklearn_version >= parse_version("1.4"):
-        # TODO: remove when the minimum supported version of scikit-learn will be 1.4
-        # support for missing values
-        params_parallel_build_trees["missing_values_in_feature_mask"] = (
-            missing_values_in_feature_mask
-        )
+    params_parallel_build_trees["missing_values_in_feature_mask"] = (
+        missing_values_in_feature_mask
+    )
 
     tree = _parallel_build_trees(**params_parallel_build_trees)
 
@@ -469,20 +466,9 @@ def __init__(
             "min_impurity_decrease": min_impurity_decrease,
             "ccp_alpha": ccp_alpha,
             "max_samples": max_samples,
+            "monotonic_cst": monotonic_cst,
         }
-        # TODO: remove when the minimum supported version of scikit-learn will be 1.4
-        if sklearn_version >= parse_version("1.4"):
-            # use scikit-learn support for monotonic constraints
-            params_random_forest["monotonic_cst"] = monotonic_cst
-        else:
-            if monotonic_cst is not None:
-                raise ValueError(
-                    "Monotonic constraints are not supported for scikit-learn "
-                    "version < 1.4."
-                )
-            # create an attribute for compatibility with other scikit-learn tools such
-            # as HTML representation.
-            self.monotonic_cst = monotonic_cst
+
         super().__init__(**params_random_forest)
 
         self.sampling_strategy = sampling_strategy
@@ -548,37 +534,26 @@ def fit(self, X, y, sample_weight=None):
         if issparse(y):
             raise ValueError("sparse multilabel-indicator for y is not supported.")
 
-        # TODO: remove when the minimum supported version of scipy will be 1.4
-        # Support for missing values
-        if sklearn_version >= parse_version("1.4"):
-            ensure_all_finite = False
-        else:
-            ensure_all_finite = True
-
         X, y = validate_data(
             self,
             X=X,
             y=y,
             multi_output=True,
             accept_sparse="csc",
             dtype=DTYPE,
-            ensure_all_finite=ensure_all_finite,
+            ensure_all_finite=False,
         )
 
-        # TODO: remove when the minimum supported version of scikit-learn will be 1.4
-        if sklearn_version >= parse_version("1.4"):
-            # _compute_missing_values_in_feature_mask checks if X has missing values and
-            # will raise an error if the underlying tree base estimator can't handle
-            # missing values. Only the criterion is required to determine if the tree
-            # supports missing values.
-            estimator = type(self.estimator)(criterion=self.criterion)
-            missing_values_in_feature_mask = (
-                estimator._compute_missing_values_in_feature_mask(
-                    X, estimator_name=self.__class__.__name__
-                )
+        # _compute_missing_values_in_feature_mask checks if X has missing values and
+        # will raise an error if the underlying tree base estimator can't handle
+        # missing values. Only the criterion is required to determine if the tree
+        # supports missing values.
+        estimator = type(self.estimator)(criterion=self.criterion)
+        missing_values_in_feature_mask = (
+            estimator._compute_missing_values_in_feature_mask(
+                X, estimator_name=self.__class__.__name__
             )
-        else:
-            missing_values_in_feature_mask = None
+        )
 
         if sample_weight is not None:
             sample_weight = _check_sample_weight(sample_weight, X)

imblearn/keras/_generator.py

@@ -20,7 +20,7 @@ def import_from_keras():
             if hasattr(keras.utils, "Sequence"):
                 return (keras.utils.Sequence,), True
             else:
-                return (keras.utils.data_utils.Sequence,), True
+                return (keras.utils.PyDataset,), True
         except ImportError:
             return tuple(), False
 
@@ -31,7 +31,7 @@ def import_from_tensforflow():
             if hasattr(keras.utils, "Sequence"):
                 return (keras.utils.Sequence,), True
             else:
-                return (keras.utils.data_utils.Sequence,), True
+                return (keras.utils.PyDataset,), True
         except ImportError:
             return tuple(), False