[enhancement] Enable Array API in ensemble algos

deselected_tests.yaml

@@ -525,12 +525,8 @@ gpu:
   - tests/test_common.py::test_estimators[ExtraTreesClassifier()-check_class_weight_classifiers]
   - tests/test_common.py::test_estimators[ExtraTreesRegressor()-check_sample_weights_invariance(kind=zeros)]
   - tests/test_common.py::test_estimators[RandomForestRegressor()-check_regressor_data_not_an_array]
-
-  # GPU implementation of Extra Trees doesn't support sample_weights
-  # comparisons to GPU with sample weights will use different algorithms
-  - tests/test_common.py::test_estimators[ExtraTreesClassifier()-check_sample_weights_invariance(kind=ones)]
   - tests/test_common.py::test_estimators[ExtraTreesClassifier()-check_sample_weights_invariance(kind=zeros)]
-  - tests/test_common.py::test_estimators[ExtraTreesRegressor()-check_sample_weights_invariance(kind=ones)]
+  - ensemble/tests/test_forest.py::test_min_weight_fraction_leaf
 
   # RuntimeError: Device support is not implemented, failing as result of fallback to cpu false
   - svm/tests/test_svm.py::test_unfitted

doc/sources/array_api.rst

@@ -97,6 +97,10 @@ The following patched classes have support for array API inputs:
 - :obj:`sklearn.covariance.EmpiricalCovariance`
 - :obj:`sklearnex.covariance.IncrementalEmpiricalCovariance`
 - :obj:`sklearn.decomposition.PCA`
+- :obj:`sklearn.ensemble.ExtraTreesClassifier`
+- :obj:`sklearn.ensemble.ExtraTreesRegressor`
+- :obj:`sklearn.ensemble.RandomForestClassifier`
+- :obj:`sklearn.ensemble.RandomForestRegressor`
 - :obj:`sklearn.linear_model.LinearRegression`
 - :obj:`sklearn.linear_model.Ridge`
 - :obj:`sklearnex.linear_model.IncrementalLinearRegression`
@@ -108,6 +112,11 @@ The following patched classes have support for array API inputs:
     enabled in |sklearn|, when passing these classes as inputs, data will be transferred to host and then back to
     device instead of being used directly.
 
+    Result attributes of |sklearnex| classes which contain |sklearn| or |sklearnex| classes may not themselves be
+    array API compliant. For example, ensemble algorithms contain decision tree estimators result objects which
+    do not comply with the array API standard.
+
+
 
 Example usage
 =============