Fix compatibility with scikit-learn 1.8+

scikit-learn 1.8.0+ removed the algorithm parameter from AdaBoostClassifier

scikit-learn/scikit-learn#32262

The changes maintain compatibility with previous version of scikit-learn

Author: stratakis

doc/ensemble.rst

@@ -97,8 +97,7 @@ Several methods taking advantage of boosting have been designed.
 a boosting iteration :cite:`seiffert2009rusboost`::
 
   >>> from imblearn.ensemble import RUSBoostClassifier
-  >>> rusboost = RUSBoostClassifier(n_estimators=200, algorithm='SAMME.R',
-  ...                               random_state=0)
+  >>> rusboost = RUSBoostClassifier(n_estimators=200, random_state=0)
   >>> rusboost.fit(X_train, y_train)
   RUSBoostClassifier(...)
   >>> y_pred = rusboost.predict(X_test)

examples/ensemble/plot_comparison_ensemble_classifier.py

@@ -194,10 +194,16 @@
 
 # %%
 from sklearn.ensemble import AdaBoostClassifier
+from sklearn.utils.fixes import parse_version
 
 from imblearn.ensemble import EasyEnsembleClassifier, RUSBoostClassifier
+from imblearn.utils._sklearn_compat import sklearn_version
+
+if sklearn_version < parse_version("1.6"):
+    estimator = AdaBoostClassifier(n_estimators=10, algorithm="SAMME")
+else:
+    estimator = AdaBoostClassifier(n_estimators=10)
 
-estimator = AdaBoostClassifier(n_estimators=10, algorithm="SAMME")
 eec = EasyEnsembleClassifier(n_estimators=10, estimator=estimator)
 eec.fit(X_train, y_train)
 y_pred_eec = eec.predict(X_test)

imblearn/ensemble/_easy_ensemble.py

@@ -226,12 +226,14 @@ def _validate_y(self, y):
             self._sampling_strategy = self.sampling_strategy
         return y_encoded
 
-    def _validate_estimator(self, default=AdaBoostClassifier(algorithm="SAMME")):
+    def _validate_estimator(self, default=None):
         """Check the estimator and the n_estimator attribute, set the
         `estimator_` attribute."""
         if self.estimator is not None:
             estimator = clone(self.estimator)
         else:
+            if default is None:
+                default = self._get_estimator()
             estimator = clone(default)
 
         sampler = RandomUnderSampler(
@@ -279,7 +281,7 @@ def base_estimator_(self):
 
     def _get_estimator(self):
         if self.estimator is None:
-            if parse_version("1.4") <= sklearn_version < parse_version("1.6"):
+            if sklearn_version < parse_version("1.6"):
                 return AdaBoostClassifier(algorithm="SAMME")
             else:
                 return AdaBoostClassifier()