scikit-learn-intelex 2024.0.1__py312-none-win_amd64.whl

scikit_learn_intelex-2024.0.1.data/data/Lib/site-packages/sklearnex/spmd/ensemble/forest.py

@@ -0,0 +1,79 @@
+# ==============================================================================
+# Copyright 2023 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.
+# ==============================================================================
+
+from abc import ABC
+
+from onedal.spmd.ensemble import RandomForestClassifier as onedal_RandomForestClassifier
+from onedal.spmd.ensemble import RandomForestRegressor as onedal_RandomForestRegressor
+
+from ...ensemble import RandomForestClassifier as RandomForestClassifier_Batch
+from ...ensemble import RandomForestRegressor as RandomForestRegressor_Batch
+
+
+class BaseForestSPMD(ABC):
+    def _onedal_classifier(self, **onedal_params):
+        return onedal_RandomForestClassifier(**onedal_params)
+
+    def _onedal_regressor(self, **onedal_params):
+        return onedal_RandomForestRegressor(**onedal_params)
+
+
+class RandomForestClassifier(BaseForestSPMD, RandomForestClassifier_Batch):
+    __doc__ = RandomForestClassifier_Batch.__doc__
+
+    def _onedal_cpu_supported(self, method_name, *data):
+        # TODO:
+        # check which methods supported SPMD interface on CPU.
+        ready = super()._onedal_cpu_supported(method_name, *data)
+        if not ready:
+            raise RuntimeError(
+                f"Method {method_name} in {self.__class__.__name__} "
+                "is not supported with given inputs."
+            )
+        return ready
+
+    def _onedal_gpu_supported(self, method_name, *data):
+        ready = super()._onedal_gpu_supported(method_name, *data)
+        if not ready:
+            raise RuntimeError(
+                f"Method {method_name} in {self.__class__.__name__} "
+                "is not supported with given inputs."
+            )
+        return ready
+
+
+class RandomForestRegressor(BaseForestSPMD, RandomForestRegressor_Batch):
+    __doc__ = RandomForestRegressor_Batch.__doc__
+
+    def _onedal_cpu_supported(self, method_name, *data):
+        # TODO:
+        # check which methods supported SPMD interface on CPU.
+        ready = super()._onedal_cpu_supported(method_name, *data)
+        if not ready:
+            raise RuntimeError(
+                f"Method {method_name} in {self.__class__.__name__} "
+                "is not supported with given inputs."
+            )
+        return ready
+
+    def _onedal_gpu_supported(self, method_name, *data):
+        ready = super()._onedal_gpu_supported(method_name, *data)
+        if not ready:
+            raise RuntimeError(
+                f"Method {method_name} in {self.__class__.__name__} "
+                "is not supported with given inputs."
+            )
+        return ready

scikit_learn_intelex-2024.0.1.data/data/Lib/site-packages/sklearnex/spmd/linear_model/__init__.py

@@ -0,0 +1,19 @@
+# ==============================================================================
+# Copyright 2023 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.
+# ==============================================================================
+
+from .linear_model import LinearRegression
+
+__all__ = ["LinearRegression"]

scikit_learn_intelex-2024.0.1.data/data/Lib/site-packages/sklearnex/spmd/linear_model/linear_model.py

@@ -0,0 +1,21 @@
+# ==============================================================================
+# Copyright 2023 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.
+# ==============================================================================
+
+from onedal.spmd.linear_model import LinearRegression
+
+# TODO:
+# Currently it uses `onedal` module interface.
+# Add sklearnex dispatching.

scikit_learn_intelex-2024.0.1.data/data/Lib/site-packages/sklearnex/spmd/neighbors/__init__.py

@@ -0,0 +1,19 @@
+# ==============================================================================
+# Copyright 2023 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.
+# ==============================================================================
+
+from .neighbors import KNeighborsClassifier, KNeighborsRegressor, NearestNeighbors
+
+__all__ = ["KNeighborsClassifier", "KNeighborsRegressor", "NearestNeighbors"]

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

@@ -0,0 +1,25 @@
+# ==============================================================================
+# Copyright 2023 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.
+# ==============================================================================
+
+from onedal.spmd.neighbors import (
+    KNeighborsClassifier,
+    KNeighborsRegressor,
+    NearestNeighbors,
+)
+
+# TODO:
+# Currently it uses `onedal` module interface.
+# Add sklearnex dispatching.

scikit_learn_intelex-2024.0.1.data/data/Lib/site-packages/sklearnex/svm/__init__.py

@@ -0,0 +1,30 @@
+#!/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.
+# ==============================================================================
+
+from .._utils import get_sklearnex_version
+
+if get_sklearnex_version((2021, "P", 300)):
+    from .nusvc import NuSVC
+    from .nusvr import NuSVR
+    from .svc import SVC
+    from .svr import SVR
+
+    __all__ = ["SVR", "SVC", "NuSVC", "NuSVR"]
+else:
+    from daal4py.sklearn.svm import SVC
+
+    __all__ = ["SVC"]

scikit_learn_intelex-2024.0.1.data/data/Lib/site-packages/sklearnex/svm/_common.py

@@ -0,0 +1,188 @@
+# ==============================================================================
+# 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.
+# ==============================================================================
+
+from abc import ABC
+
+import numpy as np
+from sklearn.calibration import CalibratedClassifierCV
+from sklearn.model_selection import StratifiedKFold
+from sklearn.preprocessing import LabelEncoder
+
+from daal4py.sklearn._utils import sklearn_check_version
+from onedal.utils import _column_or_1d
+
+from .._utils import PatchingConditionsChain
+
+
+def get_dual_coef(self):
+    return self.dual_coef_
+
+
+def set_dual_coef(self, value):
+    self.dual_coef_ = value
+    if hasattr(self, "_onedal_estimator"):
+        self._onedal_estimator.dual_coef_ = value
+        if not self._is_in_fit:
+            del self._onedal_estimator._onedal_model
+
+
+def get_intercept(self):
+    return self._intercept_
+
+
+def set_intercept(self, value):
+    self._intercept_ = value
+    if hasattr(self, "_onedal_estimator"):
+        self._onedal_estimator.intercept_ = value
+        if not self._is_in_fit:
+            del self._onedal_estimator._onedal_model
+
+
+class BaseSVM(ABC):
+    def _onedal_gpu_supported(self, method_name, *data):
+        patching_status = PatchingConditionsChain(f"sklearn.{method_name}")
+        patching_status.and_conditions([(False, "GPU offloading is not supported.")])
+        return patching_status
+
+    def _onedal_cpu_supported(self, method_name, *data):
+        class_name = self.__class__.__name__
+        patching_status = PatchingConditionsChain(
+            f"sklearn.svm.{class_name}.{method_name}"
+        )
+        if method_name == "fit":
+            patching_status.and_conditions(
+                [
+                    (
+                        self.kernel in ["linear", "rbf", "poly", "sigmoid"],
+                        f'Kernel is "{self.kernel}" while '
+                        '"linear", "rbf", "poly" and "sigmoid" are only supported.',
+                    )
+                ]
+            )
+            return patching_status
+        inference_methods = (
+            ["predict"]
+            if class_name.endswith("R")
+            else ["predict", "predict_proba", "decision_function"]
+        )
+        if method_name in inference_methods:
+            patching_status.and_conditions(
+                [(hasattr(self, "_onedal_estimator"), "oneDAL model was not trained.")]
+            )
+            return patching_status
+        raise RuntimeError(f"Unknown method {method_name} in {class_name}")
+
+
+class BaseSVC(BaseSVM):
+    def _compute_balanced_class_weight(self, y):
+        y_ = _column_or_1d(y)
+        classes, _ = np.unique(y_, return_inverse=True)
+
+        le = LabelEncoder()
+        y_ind = le.fit_transform(y_)
+        if not all(np.in1d(classes, le.classes_)):
+            raise ValueError("classes should have valid labels that are in y")
+
+        recip_freq = len(y_) / (len(le.classes_) * np.bincount(y_ind).astype(np.float64))
+        return recip_freq[le.transform(classes)]
+
+    def _fit_proba(self, X, y, sample_weight=None, queue=None):
+        params = self.get_params()
+        params["probability"] = False
+        params["decision_function_shape"] = "ovr"
+        clf_base = self.__class__(**params)
+
+        try:
+            n_splits = 5
+            n_jobs = n_splits if queue is None or queue.sycl_device.is_cpu else 1
+            cv = StratifiedKFold(
+                n_splits=n_splits, shuffle=True, random_state=self.random_state
+            )
+            if sklearn_check_version("0.24"):
+                self.clf_prob = CalibratedClassifierCV(
+                    clf_base, ensemble=False, cv=cv, method="sigmoid", n_jobs=n_jobs
+                )
+            else:
+                self.clf_prob = CalibratedClassifierCV(clf_base, cv=cv, method="sigmoid")
+            self.clf_prob.fit(X, y, sample_weight)
+        except ValueError:
+            clf_base = clf_base.fit(X, y, sample_weight)
+            self.clf_prob = CalibratedClassifierCV(
+                clf_base, cv="prefit", method="sigmoid"
+            )
+            self.clf_prob.fit(X, y, sample_weight)
+
+    def _save_attributes(self):
+        self.support_vectors_ = self._onedal_estimator.support_vectors_
+        self.n_features_in_ = self._onedal_estimator.n_features_in_
+        self.fit_status_ = 0
+        self.dual_coef_ = self._onedal_estimator.dual_coef_
+        self.shape_fit_ = self._onedal_estimator.class_weight_
+        self.classes_ = self._onedal_estimator.classes_
+        self.class_weight_ = self._onedal_estimator.class_weight_
+        self.support_ = self._onedal_estimator.support_
+
+        self._intercept_ = self._onedal_estimator.intercept_
+        self._n_support = self._onedal_estimator._n_support
+        self._sparse = False
+        self._gamma = self._onedal_estimator._gamma
+        if self.probability:
+            length = int(len(self.classes_) * (len(self.classes_) - 1) / 2)
+            self._probA = np.zeros(length)
+            self._probB = np.zeros(length)
+        else:
+            self._probA = np.empty(0)
+            self._probB = np.empty(0)
+
+        self._dual_coef_ = property(get_dual_coef, set_dual_coef)
+        self.intercept_ = property(get_intercept, set_intercept)
+
+        self._is_in_fit = True
+        self._dual_coef_ = self.dual_coef_
+        self.intercept_ = self._intercept_
+        self._is_in_fit = False
+
+        if sklearn_check_version("1.1"):
+            length = int(len(self.classes_) * (len(self.classes_) - 1) / 2)
+            self.n_iter_ = np.full((length,), self._onedal_estimator.n_iter_)
+
+
+class BaseSVR(BaseSVM):
+    def _save_attributes(self):
+        self.support_vectors_ = self._onedal_estimator.support_vectors_
+        self.n_features_in_ = self._onedal_estimator.n_features_in_
+        self.fit_status_ = 0
+        self.dual_coef_ = self._onedal_estimator.dual_coef_
+        self.shape_fit_ = self._onedal_estimator.shape_fit_
+        self.support_ = self._onedal_estimator.support_
+
+        self._intercept_ = self._onedal_estimator.intercept_
+        self._n_support = [self.support_vectors_.shape[0]]
+        self._sparse = False
+        self._gamma = self._onedal_estimator._gamma
+        self._probA = None
+        self._probB = None
+
+        self._dual_coef_ = property(get_dual_coef, set_dual_coef)
+        self.intercept_ = property(get_intercept, set_intercept)
+
+        self._is_in_fit = True
+        self._dual_coef_ = self.dual_coef_
+        self.intercept_ = self._intercept_
+        self._is_in_fit = False
+
+        if sklearn_check_version("1.1"):
+            self.n_iter_ = self._onedal_estimator.n_iter_

scikit_learn_intelex-2024.0.1.data/data/Lib/site-packages/sklearnex/svm/nusvc.py

@@ -0,0 +1,272 @@
+# ==============================================================================
+# 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.
+# ==============================================================================
+
+from sklearn.exceptions import NotFittedError
+from sklearn.svm import NuSVC as sklearn_NuSVC
+from sklearn.utils.validation import _deprecate_positional_args
+
+from daal4py.sklearn._utils import sklearn_check_version
+
+from .._device_offload import dispatch, wrap_output_data
+from ._common import BaseSVC
+
+if sklearn_check_version("1.0"):
+    from sklearn.utils.metaestimators import available_if
+
+from onedal.svm import NuSVC as onedal_NuSVC
+
+
+class NuSVC(sklearn_NuSVC, BaseSVC):
+    __doc__ = sklearn_NuSVC.__doc__
+
+    if sklearn_check_version("1.2"):
+        _parameter_constraints: dict = {**sklearn_NuSVC._parameter_constraints}
+
+    @_deprecate_positional_args
+    def __init__(
+        self,
+        *,
+        nu=0.5,
+        kernel="rbf",
+        degree=3,
+        gamma="scale",
+        coef0=0.0,
+        shrinking=True,
+        probability=False,
+        tol=1e-3,
+        cache_size=200,
+        class_weight=None,
+        verbose=False,
+        max_iter=-1,
+        decision_function_shape="ovr",
+        break_ties=False,
+        random_state=None,
+    ):
+        super().__init__(
+            nu=nu,
+            kernel=kernel,
+            degree=degree,
+            gamma=gamma,
+            coef0=coef0,
+            shrinking=shrinking,
+            probability=probability,
+            tol=tol,
+            cache_size=cache_size,
+            class_weight=class_weight,
+            verbose=verbose,
+            max_iter=max_iter,
+            decision_function_shape=decision_function_shape,
+            break_ties=break_ties,
+            random_state=random_state,
+        )
+
+    def fit(self, X, y, sample_weight=None):
+        """
+        Fit the SVM model according to the given training data.
+
+        Parameters
+        ----------
+        X : {array-like, sparse matrix} of shape (n_samples, n_features) \
+                or (n_samples, n_samples)
+            Training vectors, where `n_samples` is the number of samples
+            and `n_features` is the number of features.
+            For kernel="precomputed", the expected shape of X is
+            (n_samples, n_samples).
+
+        y : array-like of shape (n_samples,)
+            Target values (class labels in classification, real numbers in
+            regression).
+
+        sample_weight : array-like of shape (n_samples,), default=None
+            Per-sample weights. Rescale C per sample. Higher weights
+            force the classifier to put more emphasis on these points.
+
+        Returns
+        -------
+        self : object
+            Fitted estimator.
+
+        Notes
+        -----
+        If X and y are not C-ordered and contiguous arrays of np.float64 and
+        X is not a scipy.sparse.csr_matrix, X and/or y may be copied.
+
+        If X is a dense array, then the other methods will not support sparse
+        matrices as input.
+        """
+        if sklearn_check_version("1.2"):
+            self._validate_params()
+        if sklearn_check_version("1.0"):
+            self._check_feature_names(X, reset=True)
+        dispatch(
+            self,
+            "fit",
+            {
+                "onedal": self.__class__._onedal_fit,
+                "sklearn": sklearn_NuSVC.fit,
+            },
+            X,
+            y,
+            sample_weight,
+        )
+
+        return self
+
+    @wrap_output_data
+    def predict(self, X):
+        """
+        Perform regression on samples in X.
+
+        For an one-class model, +1 (inlier) or -1 (outlier) is returned.
+
+        Parameters
+        ----------
+        X : {array-like, sparse matrix} of shape (n_samples, n_features)
+            For kernel="precomputed", the expected shape of X is
+            (n_samples_test, n_samples_train).
+
+        Returns
+        -------
+        y_pred : ndarray of shape (n_samples,)
+            The predicted values.
+        """
+        if sklearn_check_version("1.0"):
+            self._check_feature_names(X, reset=False)
+        return dispatch(
+            self,
+            "predict",
+            {
+                "onedal": self.__class__._onedal_predict,
+                "sklearn": sklearn_NuSVC.predict,
+            },
+            X,
+        )
+
+    if sklearn_check_version("1.0"):
+
+        @available_if(sklearn_NuSVC._check_proba)
+        def predict_proba(self, X):
+            """
+            Compute probabilities of possible outcomes for samples in X.
+
+            The model need to have probability information computed at training
+            time: fit with attribute `probability` set to True.
+
+            Parameters
+            ----------
+            X : array-like of shape (n_samples, n_features)
+                For kernel="precomputed", the expected shape of X is
+                (n_samples_test, n_samples_train).
+
+            Returns
+            -------
+            T : ndarray of shape (n_samples, n_classes)
+                Returns the probability of the sample for each class in
+                the model. The columns correspond to the classes in sorted
+                order, as they appear in the attribute :term:`classes_`.
+
+            Notes
+            -----
+            The probability model is created using cross validation, so
+            the results can be slightly different than those obtained by
+            predict. Also, it will produce meaningless results on very small
+            datasets.
+            """
+            return self._predict_proba(X)
+
+    else:
+
+        @property
+        def predict_proba(self):
+            self._check_proba()
+            return self._predict_proba
+
+    @wrap_output_data
+    def _predict_proba(self, X):
+        if sklearn_check_version("1.0"):
+            self._check_feature_names(X, reset=False)
+        sklearn_pred_proba = (
+            sklearn_NuSVC.predict_proba
+            if sklearn_check_version("1.0")
+            else sklearn_NuSVC._predict_proba
+        )
+
+        return dispatch(
+            self,
+            "predict_proba",
+            {
+                "onedal": self.__class__._onedal_predict_proba,
+                "sklearn": sklearn_pred_proba,
+            },
+            X,
+        )
+
+    @wrap_output_data
+    def decision_function(self, X):
+        if sklearn_check_version("1.0"):
+            self._check_feature_names(X, reset=False)
+        return dispatch(
+            self,
+            "decision_function",
+            {
+                "onedal": self.__class__._onedal_decision_function,
+                "sklearn": sklearn_NuSVC.decision_function,
+            },
+            X,
+        )
+
+    def _onedal_fit(self, X, y, sample_weight=None, queue=None):
+        onedal_params = {
+            "nu": self.nu,
+            "kernel": self.kernel,
+            "degree": self.degree,
+            "gamma": self.gamma,
+            "coef0": self.coef0,
+            "tol": self.tol,
+            "shrinking": self.shrinking,
+            "cache_size": self.cache_size,
+            "max_iter": self.max_iter,
+            "class_weight": self.class_weight,
+            "break_ties": self.break_ties,
+            "decision_function_shape": self.decision_function_shape,
+        }
+
+        self._onedal_estimator = onedal_NuSVC(**onedal_params)
+        self._onedal_estimator.fit(X, y, sample_weight, queue=queue)
+
+        if self.probability:
+            self._fit_proba(X, y, sample_weight, queue=queue)
+        self._save_attributes()
+
+    def _onedal_predict(self, X, queue=None):
+        return self._onedal_estimator.predict(X, queue=queue)
+
+    def _onedal_predict_proba(self, X, queue=None):
+        if getattr(self, "clf_prob", None) is None:
+            raise NotFittedError(
+                "predict_proba is not available when fitted with probability=False"
+            )
+        from .._config import config_context, get_config
+
+        # We use stock metaestimators below, so the only way
+        # to pass a queue is using config_context.
+        cfg = get_config()
+        cfg["target_offload"] = queue
+        with config_context(**cfg):
+            return self.clf_prob.predict_proba(X)
+
+    def _onedal_decision_function(self, X, queue=None):
+        return self._onedal_estimator.decision_function(X, queue=queue)