scikit-learn-intelex 2023.2.1__py38-none-win_amd64.whl → 2024.0.1__py38-none-win_amd64.whl

{scikit_learn_intelex-2023.2.1.data → scikit_learn_intelex-2024.0.1.data}/data/Lib/site-packages/sklearnex/neighbors/knn_unsupervised.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-#===============================================================================
+# ===============================================================================
 # Copyright 2021 Intel Corporation
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
@@ -13,85 +13,128 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-#===============================================================================
+# ===============================================================================
 
 try:
     from packaging.version import Version
 except ImportError:
     from distutils.version import LooseVersion as Version
-from sklearn import __version__ as sklearn_version
-from daal4py.sklearn._utils import sklearn_check_version
+
 import warnings
 
-from sklearn.neighbors._base import NeighborsBase as sklearn_NeighborsBase
+import numpy as np
+from sklearn import __version__ as sklearn_version
 from sklearn.neighbors._ball_tree import BallTree
-from sklearn.neighbors._kd_tree import KDTree
 from sklearn.neighbors._base import VALID_METRICS
-from sklearn.neighbors._unsupervised import NearestNeighbors as \
-    sklearn_NearestNeighbors
-
+from sklearn.neighbors._base import NeighborsBase as sklearn_NeighborsBase
+from sklearn.neighbors._kd_tree import KDTree
+from sklearn.neighbors._unsupervised import NearestNeighbors as sklearn_NearestNeighbors
 from sklearn.utils.validation import _deprecate_positional_args, check_is_fitted
 
-from onedal.datatypes import _check_array, _num_features, _num_samples
+from daal4py.sklearn._utils import sklearn_check_version
 from onedal.neighbors import NearestNeighbors as onedal_NearestNeighbors
+from onedal.utils import _check_array, _num_features, _num_samples
 
-from .common import KNeighborsDispatchingBase
 from .._device_offload import dispatch, wrap_output_data
-import numpy as np
+from .common import KNeighborsDispatchingBase
 
+if sklearn_check_version("0.22") and Version(sklearn_version) < Version("0.23"):
 
-if sklearn_check_version("0.22") and \
-   Version(sklearn_version) < Version("0.23"):
     class NearestNeighbors_(sklearn_NearestNeighbors):
-        def __init__(self, n_neighbors=5, radius=1.0,
-                     algorithm='auto', leaf_size=30, metric='minkowski',
-                     p=2, metric_params=None, n_jobs=None):
+        def __init__(
+            self,
+            n_neighbors=5,
+            radius=1.0,
+            algorithm="auto",
+            leaf_size=30,
+            metric="minkowski",
+            p=2,
+            metric_params=None,
+            n_jobs=None,
+        ):
             super().__init__(
                 n_neighbors=n_neighbors,
                 radius=radius,
                 algorithm=algorithm,
-                leaf_size=leaf_size, metric=metric, p=p,
-                metric_params=metric_params, n_jobs=n_jobs)
+                leaf_size=leaf_size,
+                metric=metric,
+                p=p,
+                metric_params=metric_params,
+                n_jobs=n_jobs,
+            )
+
 else:
+
     class NearestNeighbors_(sklearn_NearestNeighbors):
-        if sklearn_check_version('1.2'):
+        if sklearn_check_version("1.2"):
             _parameter_constraints: dict = {
-                **sklearn_NearestNeighbors._parameter_constraints}
+                **sklearn_NearestNeighbors._parameter_constraints
+            }
 
         @_deprecate_positional_args
-        def __init__(self, *, n_neighbors=5, radius=1.0,
-                     algorithm='auto', leaf_size=30, metric='minkowski',
-                     p=2, metric_params=None, n_jobs=None):
+        def __init__(
+            self,
+            *,
+            n_neighbors=5,
+            radius=1.0,
+            algorithm="auto",
+            leaf_size=30,
+            metric="minkowski",
+            p=2,
+            metric_params=None,
+            n_jobs=None,
+        ):
             super().__init__(
                 n_neighbors=n_neighbors,
                 radius=radius,
                 algorithm=algorithm,
-                leaf_size=leaf_size, metric=metric, p=p,
-                metric_params=metric_params, n_jobs=n_jobs)
+                leaf_size=leaf_size,
+                metric=metric,
+                p=p,
+                metric_params=metric_params,
+                n_jobs=n_jobs,
+            )
 
 
 class NearestNeighbors(NearestNeighbors_, KNeighborsDispatchingBase):
-    if sklearn_check_version('1.2'):
-        _parameter_constraints: dict = {
-            **NearestNeighbors_._parameter_constraints}
+    if sklearn_check_version("1.2"):
+        _parameter_constraints: dict = {**NearestNeighbors_._parameter_constraints}
 
     @_deprecate_positional_args
-    def __init__(self, n_neighbors=5, radius=1.0,
-                 algorithm='auto', leaf_size=30, metric='minkowski',
-                 p=2, metric_params=None, n_jobs=None):
+    def __init__(
+        self,
+        n_neighbors=5,
+        radius=1.0,
+        algorithm="auto",
+        leaf_size=30,
+        metric="minkowski",
+        p=2,
+        metric_params=None,
+        n_jobs=None,
+    ):
         super().__init__(
             n_neighbors=n_neighbors,
             radius=radius,
             algorithm=algorithm,
-            leaf_size=leaf_size, metric=metric, p=p,
-            metric_params=metric_params, n_jobs=n_jobs)
+            leaf_size=leaf_size,
+            metric=metric,
+            p=p,
+            metric_params=metric_params,
+            n_jobs=n_jobs,
+        )
 
     def fit(self, X, y=None):
         self._fit_validation(X, y)
-        dispatch(self, 'fit', {
-            'onedal': self.__class__._onedal_fit,
-            'sklearn': sklearn_NearestNeighbors.fit,
-        }, X, None)
+        dispatch(
+            self,
+            "fit",
+            {
+                "onedal": self.__class__._onedal_fit,
+                "sklearn": sklearn_NearestNeighbors.fit,
+            },
+            X,
+            None,
+        )
         return self
 
     @wrap_output_data
@@ -99,37 +142,50 @@ class NearestNeighbors(NearestNeighbors_, KNeighborsDispatchingBase):
         check_is_fitted(self)
         if sklearn_check_version("1.0") and X is not None:
             self._check_feature_names(X, reset=False)
-        return dispatch(self, 'kneighbors', {
-            'onedal': self.__class__._onedal_kneighbors,
-            'sklearn': sklearn_NearestNeighbors.kneighbors,
-        }, X, n_neighbors, return_distance)
+        return dispatch(
+            self,
+            "kneighbors",
+            {
+                "onedal": self.__class__._onedal_kneighbors,
+                "sklearn": sklearn_NearestNeighbors.kneighbors,
+            },
+            X,
+            n_neighbors,
+            return_distance,
+        )
 
     @wrap_output_data
-    def radius_neighbors(self, X=None, radius=None, return_distance=True,
-                         sort_results=False):
-        _onedal_estimator = getattr(self, '_onedal_estimator', None)
-
-        if _onedal_estimator is not None or getattr(self, '_tree', 0) is None and \
-                self._fit_method == 'kd_tree':
+    def radius_neighbors(
+        self, X=None, radius=None, return_distance=True, sort_results=False
+    ):
+        _onedal_estimator = getattr(self, "_onedal_estimator", None)
+
+        if (
+            _onedal_estimator is not None
+            or getattr(self, "_tree", 0) is None
+            and self._fit_method == "kd_tree"
+        ):
             if sklearn_check_version("0.24"):
-                sklearn_NearestNeighbors.fit(self, self._fit_X, getattr(self, '_y', None))
+                sklearn_NearestNeighbors.fit(self, self._fit_X, getattr(self, "_y", None))
             else:
                 sklearn_NearestNeighbors.fit(self, self._fit_X)
         if sklearn_check_version("0.22"):
             result = sklearn_NearestNeighbors.radius_neighbors(
-                self, X, radius, return_distance, sort_results)
+                self, X, radius, return_distance, sort_results
+            )
         else:
             result = sklearn_NearestNeighbors.radius_neighbors(
-                self, X, radius, return_distance)
+                self, X, radius, return_distance
+            )
 
         return result
 
     def _onedal_fit(self, X, y=None, queue=None):
         onedal_params = {
-            'n_neighbors': self.n_neighbors,
-            'algorithm': self.algorithm,
-            'metric': self.effective_metric_,
-            'p': self.effective_metric_params_['p'],
+            "n_neighbors": self.n_neighbors,
+            "algorithm": self.algorithm,
+            "metric": self.effective_metric_,
+            "p": self.effective_metric_params_["p"],
         }
 
         try:
@@ -148,10 +204,12 @@ class NearestNeighbors(NearestNeighbors_, KNeighborsDispatchingBase):
     def _onedal_predict(self, X, queue=None):
         return self._onedal_estimator.predict(X, queue=queue)
 
-    def _onedal_kneighbors(self, X=None, n_neighbors=None,
-                           return_distance=True, queue=None):
+    def _onedal_kneighbors(
+        self, X=None, n_neighbors=None, return_distance=True, queue=None
+    ):
         return self._onedal_estimator.kneighbors(
-            X, n_neighbors, return_distance, queue=queue)
+            X, n_neighbors, return_distance, queue=queue
+        )
 
     def _save_attributes(self):
         self.classes_ = self._onedal_estimator.classes_

{scikit_learn_intelex-2023.2.1.data → scikit_learn_intelex-2024.0.1.data}/data/Lib/site-packages/sklearnex/neighbors/lof.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-#===============================================================================
+# ===============================================================================
 # Copyright 2023 Intel Corporation
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
@@ -13,13 +13,13 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-#===============================================================================
+# ===============================================================================
 
-import numpy as np
 import warnings
 
-from sklearn.neighbors._lof import LocalOutlierFactor as \
-    sklearn_LocalOutlierFactor
+import numpy as np
+from sklearn.neighbors._lof import LocalOutlierFactor as sklearn_LocalOutlierFactor
+
 from .knn_unsupervised import NearestNeighbors
 
 try:
@@ -27,18 +27,22 @@ try:
 except ImportError:
     pass
 
-from sklearn.utils.validation import check_is_fitted
 from sklearn.utils import check_array
+from sklearn.utils.validation import check_is_fitted
 
 from daal4py.sklearn._utils import sklearn_check_version
-from .._device_offload import dispatch, wrap_output_data
+
 from .._config import config_context
+from .._device_offload import dispatch, wrap_output_data
+from .._utils import PatchingConditionsChain
 
 if sklearn_check_version("1.0"):
+
     class LocalOutlierFactor(sklearn_LocalOutlierFactor):
-        if sklearn_check_version('1.2'):
+        if sklearn_check_version("1.2"):
             _parameter_constraints: dict = {
-                **sklearn_LocalOutlierFactor._parameter_constraints}
+                **sklearn_LocalOutlierFactor._parameter_constraints
+            }
 
         def __init__(
             self,
@@ -62,7 +66,7 @@ if sklearn_check_version("1.0"):
                 metric_params=metric_params,
                 n_jobs=n_jobs,
                 contamination=contamination,
-                novelty=novelty
+                novelty=novelty,
             )
 
         def _fit(self, X, y, queue=None):
@@ -76,7 +80,7 @@ if sklearn_check_version("1.0"):
                     metric=self.metric,
                     p=self.p,
                     metric_params=self.metric_params,
-                    n_jobs=self.n_jobs
+                    n_jobs=self.n_jobs,
                 )
                 self._knn.fit(X)
 
@@ -98,8 +102,9 @@ if sklearn_check_version("1.0"):
                     )
                 self.n_neighbors_ = max(1, min(self.n_neighbors, n_samples - 1))
 
-                self._distances_fit_X_, _neighbors_indices_fit_X_ =\
-                    self._knn.kneighbors(n_neighbors=self.n_neighbors_)
+                self._distances_fit_X_, _neighbors_indices_fit_X_ = self._knn.kneighbors(
+                    n_neighbors=self.n_neighbors_
+                )
 
                 self._lrd = self._local_reachability_density(
                     self._distances_fit_X_, _neighbors_indices_fit_X_
@@ -127,10 +132,16 @@ if sklearn_check_version("1.0"):
                 return self
 
         def fit(self, X, y=None):
-            return dispatch(self, 'neighbors.LocalOutlierFactor.fit', {
-                'onedal': self.__class__._fit,
-                'sklearn': None,
-            }, X, y)
+            return dispatch(
+                self,
+                "neighbors.LocalOutlierFactor.fit",
+                {
+                    "onedal": self.__class__._fit,
+                    "sklearn": None,
+                },
+                X,
+                y,
+            )
 
         def _onedal_predict(self, X, queue=None):
             with config_context(target_offload=queue):
@@ -148,10 +159,15 @@ if sklearn_check_version("1.0"):
 
         @wrap_output_data
         def _predict(self, X=None):
-            return dispatch(self, 'neighbors.LocalOutlierFactor.predict', {
-                'onedal': self.__class__._onedal_predict,
-                'sklearn': None,
-            }, X)
+            return dispatch(
+                self,
+                "neighbors.LocalOutlierFactor.predict",
+                {
+                    "onedal": self.__class__._onedal_predict,
+                    "sklearn": None,
+                },
+                X,
+            )
 
         def _score_samples(self, X, queue=None):
             with config_context(target_offload=queue):
@@ -183,10 +199,15 @@ if sklearn_check_version("1.0"):
         @available_if(_check_novelty_score_samples)
         @wrap_output_data
         def score_samples(self, X):
-            return dispatch(self, 'neighbors.LocalOutlierFactor.score_samples', {
-                'onedal': self.__class__._score_samples,
-                'sklearn': None,
-            }, X)
+            return dispatch(
+                self,
+                "neighbors.LocalOutlierFactor.score_samples",
+                {
+                    "onedal": self.__class__._score_samples,
+                    "sklearn": None,
+                },
+                X,
+            )
 
         def _check_novelty_fit_predict(self):
             if self.novelty:
@@ -204,17 +225,33 @@ if sklearn_check_version("1.0"):
         @available_if(_check_novelty_fit_predict)
         @wrap_output_data
         def fit_predict(self, X, y=None):
-            return dispatch(self, 'neighbors.LocalOutlierFactor.fit_predict', {
-                'onedal': self.__class__._fit_predict,
-                'sklearn': None,
-            }, X, y)
+            return dispatch(
+                self,
+                "neighbors.LocalOutlierFactor.fit_predict",
+                {
+                    "onedal": self.__class__._fit_predict,
+                    "sklearn": None,
+                },
+                X,
+                y,
+            )
 
         def _onedal_gpu_supported(self, method_name, *data):
-            return True
+            class_name = self.__class__.__name__
+            patching_status = PatchingConditionsChain(
+                f"sklearn.neighbors.{class_name}.{method_name}"
+            )
+            return patching_status
 
         def _onedal_cpu_supported(self, method_name, *data):
-            return True
+            class_name = self.__class__.__name__
+            patching_status = PatchingConditionsChain(
+                f"sklearn.neighbors.{class_name}.{method_name}"
+            )
+            return patching_status
+
 else:
+
     class LocalOutlierFactor(sklearn_LocalOutlierFactor):
         def __init__(
             self,
@@ -238,7 +275,7 @@ else:
                 metric_params=metric_params,
                 n_jobs=n_jobs,
                 contamination=contamination,
-                novelty=novelty
+                novelty=novelty,
             )
 
         def _fit(self, X, y=None, queue=None):
@@ -250,7 +287,7 @@ else:
                     metric=self.metric,
                     p=self.p,
                     metric_params=self.metric_params,
-                    n_jobs=self.n_jobs
+                    n_jobs=self.n_jobs,
                 )
                 self._knn.fit(X)
 
@@ -272,8 +309,9 @@ else:
                     )
                 self.n_neighbors_ = max(1, min(self.n_neighbors, n_samples - 1))
 
-                self._distances_fit_X_, _neighbors_indices_fit_X_ =\
-                    self._knn.kneighbors(n_neighbors=self.n_neighbors_)
+                self._distances_fit_X_, _neighbors_indices_fit_X_ = self._knn.kneighbors(
+                    n_neighbors=self.n_neighbors_
+                )
 
                 self._lrd = self._local_reachability_density(
                     self._distances_fit_X_, _neighbors_indices_fit_X_
@@ -301,10 +339,16 @@ else:
                 return self
 
         def fit(self, X, y=None):
-            return dispatch(self, 'neighbors.LocalOutlierFactor.fit', {
-                'onedal': self.__class__._fit,
-                'sklearn': None,
-            }, X, y)
+            return dispatch(
+                self,
+                "neighbors.LocalOutlierFactor.fit",
+                {
+                    "onedal": self.__class__._fit,
+                    "sklearn": None,
+                },
+                X,
+                y,
+            )
 
         def _onedal_predict(self, X, queue=None):
             with config_context(target_offload=queue):
@@ -322,10 +366,15 @@ else:
 
         @wrap_output_data
         def _predict(self, X=None):
-            return dispatch(self, 'neighbors.LocalOutlierFactor.predict', {
-                'onedal': self.__class__._onedal_predict,
-                'sklearn': None,
-            }, X)
+            return dispatch(
+                self,
+                "neighbors.LocalOutlierFactor.predict",
+                {
+                    "onedal": self.__class__._onedal_predict,
+                    "sklearn": None,
+                },
+                X,
+            )
 
         def _onedal_score_samples(self, X, queue=None):
             with config_context(target_offload=queue):
@@ -345,17 +394,24 @@ else:
         @wrap_output_data
         def _score_samples(self, X):
             if not self.novelty:
-                msg = ('score_samples is not available when novelty=False. The '
-                       'scores of the training samples are always available '
-                       'through the negative_outlier_factor_ attribute. Use '
-                       'novelty=True if you want to use LOF for novelty detection '
-                       'and compute score_samples for new unseen data.')
+                msg = (
+                    "score_samples is not available when novelty=False. The "
+                    "scores of the training samples are always available "
+                    "through the negative_outlier_factor_ attribute. Use "
+                    "novelty=True if you want to use LOF for novelty detection "
+                    "and compute score_samples for new unseen data."
+                )
                 raise AttributeError(msg)
 
-            return dispatch(self, 'neighbors.LocalOutlierFactor.score_samples', {
-                'onedal': self.__class__._onedal_score_samples,
-                'sklearn': None,
-            }, X)
+            return dispatch(
+                self,
+                "neighbors.LocalOutlierFactor.score_samples",
+                {
+                    "onedal": self.__class__._onedal_score_samples,
+                    "sklearn": None,
+                },
+                X,
+            )
 
         def _onedal_fit_predict(self, X, y, queue=None):
             with config_context(target_offload=queue):
@@ -363,10 +419,16 @@ else:
 
         @wrap_output_data
         def _fit_predict(self, X, y=None):
-            return dispatch(self, 'neighbors.LocalOutlierFactor._onedal_fit_predict', {
-                'onedal': self.__class__._onedal_fit_predict,
-                'sklearn': None,
-            }, X, y)
+            return dispatch(
+                self,
+                "neighbors.LocalOutlierFactor._onedal_fit_predict",
+                {
+                    "onedal": self.__class__._onedal_fit_predict,
+                    "sklearn": None,
+                },
+                X,
+                y,
+            )
 
         def _onedal_gpu_supported(self, method_name, *data):
             return True

scikit_learn_intelex-2024.0.1.data/data/Lib/site-packages/sklearnex/neighbors/tests/test_neighbors.py

@@ -0,0 +1,85 @@
+#!/usr/bin/env python
+# ===============================================================================
+# Copyright 2021 Intel Corporation
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ===============================================================================
+
+import numpy as np
+import pytest
+from numpy.testing import assert_allclose
+
+from onedal.tests.utils._dataframes_support import (
+    _as_numpy,
+    _convert_to_dataframe,
+    get_dataframes_and_queues,
+)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+def test_sklearnex_import_knn_classifier(dataframe, queue):
+    from sklearnex.neighbors import KNeighborsClassifier
+
+    X = _convert_to_dataframe([[0], [1], [2], [3]], sycl_queue=queue, target_df=dataframe)
+    y = _convert_to_dataframe([0, 0, 1, 1], sycl_queue=queue, target_df=dataframe)
+    neigh = KNeighborsClassifier(n_neighbors=3).fit(X, y)
+    y_test = _convert_to_dataframe([[1.1]], sycl_queue=queue, target_df=dataframe)
+    pred = _as_numpy(neigh.predict(y_test))
+    assert "sklearnex" in neigh.__module__
+    assert_allclose(pred, [0])
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+def test_sklearnex_import_knn_regression(dataframe, queue):
+    from sklearnex.neighbors import KNeighborsRegressor
+
+    X = _convert_to_dataframe([[0], [1], [2], [3]], sycl_queue=queue, target_df=dataframe)
+    y = _convert_to_dataframe([0, 0, 1, 1], sycl_queue=queue, target_df=dataframe)
+    neigh = KNeighborsRegressor(n_neighbors=2).fit(X, y)
+    y_test = _convert_to_dataframe([[1.5]], sycl_queue=queue, target_df=dataframe)
+    pred = _as_numpy(neigh.predict(y_test))
+    assert "sklearnex" in neigh.__module__
+    assert_allclose(pred, [0.5])
+
+
+# TODO:
+# investigate failure for `dpnp.ndarrays` and `dpctl.tensors`.
+@pytest.mark.parametrize(
+    "dataframe,queue", get_dataframes_and_queues(dataframe_filter_="numpy")
+)
+def test_sklearnex_import_nn(dataframe, queue):
+    from sklearnex.neighbors import NearestNeighbors
+
+    X = [[0, 0, 2], [1, 0, 0], [0, 0, 1]]
+    X = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    test = _convert_to_dataframe([[0, 0, 1.3]], sycl_queue=queue, target_df=dataframe)
+    neigh = NearestNeighbors(n_neighbors=2).fit(X)
+    result = neigh.kneighbors(test, 2, return_distance=False)
+    result = _as_numpy(result)
+    assert "sklearnex" in neigh.__module__
+    assert_allclose(result, [[2, 0]])
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+def test_sklearnex_import_lof(dataframe, queue):
+    from sklearnex.neighbors import LocalOutlierFactor
+
+    X = [[7, 7, 7], [1, 0, 0], [0, 0, 1], [0, 0, 1]]
+    X = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    lof = LocalOutlierFactor(n_neighbors=2)
+    result = lof.fit_predict(X)
+    result = _as_numpy(result)
+    assert hasattr(lof, "_knn")
+    assert "sklearnex" in lof.__module__
+    assert "sklearnex" in lof._knn.__module__
+    assert_allclose(result, [-1, 1, 1, 1])