scikit-learn-intelex 2024.4.0__py312-none-win_amd64.whl → 2025.10.0__py312-none-win_amd64.whl

scikit_learn_intelex-2024.4.0.data/data/Lib/site-packages/sklearnex/tests/test_n_jobs_support.py

@@ -1,99 +0,0 @@
-# ==============================================================================
-# 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.
-# ==============================================================================
-
-import inspect
-import logging
-from multiprocessing import cpu_count
-
-import pytest
-from sklearn.base import BaseEstimator
-from sklearn.datasets import make_classification
-
-from sklearnex.dispatcher import get_patch_map
-from sklearnex.svm import SVC, NuSVC
-
-ESTIMATORS = set(
-    filter(
-        lambda x: inspect.isclass(x) and issubclass(x, BaseEstimator),
-        [value[0][0][2] for value in get_patch_map().values()],
-    )
-)
-
-X, Y = make_classification(n_samples=40, n_features=4, random_state=42)
-
-
-@pytest.mark.parametrize("estimator_class", ESTIMATORS)
-@pytest.mark.parametrize("n_jobs", [None, -1, 1, 2])
-def test_n_jobs_support(caplog, estimator_class, n_jobs):
-    def check_estimator_doc(estimator):
-        if estimator.__doc__ is not None:
-            assert "n_jobs" in estimator.__doc__
-
-    def check_n_jobs_entry_in_logs(caplog, function_name, n_jobs):
-        for rec in caplog.records:
-            if function_name in rec.message and "threads" in rec.message:
-                expected_n_jobs = n_jobs if n_jobs > 0 else cpu_count() + 1 + n_jobs
-                logging.info(f"{function_name}: setting {expected_n_jobs} threads")
-                if f"{function_name}: setting {expected_n_jobs} threads" in rec.message:
-                    return True
-        # False if n_jobs is set and not found in logs
-        return n_jobs is None
-
-    def check_method(*args, method, caplog):
-        method(*args)
-        assert check_n_jobs_entry_in_logs(caplog, method.__name__, n_jobs)
-
-    def check_methods_decoration(estimator):
-        funcs = {
-            i: getattr(estimator, i)
-            for i in dir(estimator)
-            if hasattr(estimator, i) and callable(getattr(estimator, i))
-        }
-
-        for func_name, func in funcs.items():
-            assert hasattr(func, "__onedal_n_jobs_decorated__") == (
-                func_name in estimator._n_jobs_supported_onedal_methods
-            ), f"{estimator}.{func_name} n_jobs decoration does not match {estimator} n_jobs supported methods"
-
-    caplog.set_level(logging.DEBUG, logger="sklearnex")
-    estimator_kwargs = {"n_jobs": n_jobs}
-    # by default, [Nu]SVC.predict_proba is restricted by @available_if decorator
-    if estimator_class in [SVC, NuSVC]:
-        estimator_kwargs["probability"] = True
-    estimator_instance = estimator_class(**estimator_kwargs)
-    # check `n_jobs` parameter doc entry
-    check_estimator_doc(estimator_class)
-    check_estimator_doc(estimator_instance)
-    # check `n_jobs` log entry for supported methods
-    # `fit` call is required before other methods
-    check_method(X, Y, method=estimator_instance.fit, caplog=caplog)
-    for method_name in estimator_instance._n_jobs_supported_onedal_methods:
-        if method_name == "fit":
-            continue
-        method = getattr(estimator_instance, method_name)
-        argdict = inspect.signature(method).parameters
-        argnum = len(
-            [i for i in argdict if argdict[i].default == inspect.Parameter.empty]
-        )
-        if argnum == 0:
-            check_method(method=method, caplog=caplog)
-        elif argnum == 1:
-            check_method(X, method=method, caplog=caplog)
-        else:
-            check_method(X, Y, method=method, caplog=caplog)
-    # check if correct methods were decorated
-    check_methods_decoration(estimator_class)
-    check_methods_decoration(estimator_instance)

scikit_learn_intelex-2024.4.0.data/data/Lib/site-packages/sklearnex/tests/test_run_to_run_stability_tests.py

@@ -1,428 +0,0 @@
-# ===============================================================================
-# Copyright 2020 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 random
-
-import numpy as np
-import pytest
-
-import daal4py as d4p
-from sklearnex import patch_sklearn
-
-patch_sklearn()
-
-from scipy import sparse
-from sklearn.cluster import DBSCAN, KMeans
-from sklearn.datasets import (
-    load_breast_cancer,
-    load_diabetes,
-    load_iris,
-    make_classification,
-    make_regression,
-)
-from sklearn.decomposition import PCA
-from sklearn.ensemble import RandomForestClassifier, RandomForestRegressor
-from sklearn.linear_model import (
-    ElasticNet,
-    Lasso,
-    LinearRegression,
-    LogisticRegression,
-    LogisticRegressionCV,
-    Ridge,
-)
-from sklearn.manifold import TSNE
-from sklearn.metrics import pairwise_distances, roc_auc_score
-from sklearn.model_selection import train_test_split
-from sklearn.neighbors import (
-    KNeighborsClassifier,
-    KNeighborsRegressor,
-    LocalOutlierFactor,
-    NearestNeighbors,
-)
-from sklearn.svm import SVC, SVR, NuSVC, NuSVR
-
-from daal4py.sklearn._utils import daal_check_version
-
-# to reproduce errors even in CI
-d4p.daalinit(nthreads=100)
-
-
-def get_class_name(x):
-    return x.__class__.__name__
-
-
-def method_processing(X, clf, methods):
-    res = []
-    name = []
-    for i in methods:
-        if i == "predict":
-            res.append(clf.predict(X))
-            name.append(get_class_name(clf) + ".predict(X)")
-        elif i == "predict_proba":
-            res.append(clf.predict_proba(X))
-            name.append(get_class_name(clf) + ".predict_proba(X)")
-        elif i == "decision_function":
-            res.append(clf.decision_function(X))
-            name.append(get_class_name(clf) + ".decision_function(X)")
-        elif i == "kneighbors":
-            dist, idx = clf.kneighbors(X)
-            res.append(dist)
-            name.append("dist")
-            res.append(idx)
-            name.append("idx")
-        elif i == "fit_predict":
-            predict = clf.fit_predict(X)
-            res.append(predict)
-            name.append(get_class_name(clf) + ".fit_predict")
-        elif i == "fit_transform":
-            res.append(clf.fit_transform(X))
-            name.append(get_class_name(clf) + ".fit_transform")
-        elif i == "transform":
-            res.append(clf.transform(X))
-            name.append(get_class_name(clf) + ".transform(X)")
-        elif i == "get_covariance":
-            res.append(clf.get_covariance())
-            name.append(get_class_name(clf) + ".get_covariance()")
-        elif i == "get_precision":
-            res.append(clf.get_precision())
-            name.append(get_class_name(clf) + ".get_precision()")
-        elif i == "score_samples":
-            res.append(clf.score_samples(X))
-            name.append(get_class_name(clf) + ".score_samples(X)")
-    return res, name
-
-
-def func(X, Y, clf, methods):
-    clf.fit(X, Y)
-    res, name = method_processing(X, clf, methods)
-
-    for i in clf.__dict__.keys():
-        ans = getattr(clf, i)
-        if isinstance(ans, (bool, float, int, np.ndarray, np.float64)):
-            if isinstance(ans, np.ndarray) and None in ans:
-                continue
-            res.append(ans)
-            name.append(get_class_name(clf) + "." + i)
-    return res, name
-
-
-def _run_test(model, methods, dataset):
-    datasets = []
-    if dataset in ["blobs", "classifier", "sparse"]:
-        X1, y1 = load_iris(return_X_y=True)
-        if dataset == "sparse":
-            X1 = sparse.csr_matrix(X1)
-        datasets.append((X1, y1))
-        X2, y2 = load_breast_cancer(return_X_y=True)
-        if dataset == "sparse":
-            X2 = sparse.csr_matrix(X2)
-        datasets.append((X2, y2))
-    elif dataset == "regression":
-        X1, y1 = make_regression(
-            n_samples=500, n_features=10, noise=64.0, random_state=42
-        )
-        datasets.append((X1, y1))
-        X2, y2 = load_diabetes(return_X_y=True)
-        datasets.append((X2, y2))
-    else:
-        raise ValueError("Unknown dataset type")
-
-    for X, y in datasets:
-        baseline, name = func(X, y, model, methods)
-        for i in range(10):
-            res, _ = func(X, y, model, methods)
-
-            for a, b, n in zip(res, baseline, name):
-                np.testing.assert_allclose(
-                    a, b, rtol=0.0, atol=0.0, err_msg=str(n + " is incorrect")
-                )
-
-
-MODELS_INFO = [
-    {
-        "model": KNeighborsClassifier(
-            n_neighbors=10, algorithm="brute", weights="uniform"
-        ),
-        "methods": ["predict", "predict_proba", "kneighbors"],
-        "dataset": "classifier",
-    },
-    {
-        "model": KNeighborsClassifier(
-            n_neighbors=10, algorithm="brute", weights="distance"
-        ),
-        "methods": ["predict", "predict_proba", "kneighbors"],
-        "dataset": "classifier",
-    },
-    {
-        "model": KNeighborsClassifier(
-            n_neighbors=10, algorithm="kd_tree", weights="uniform"
-        ),
-        "methods": ["predict", "predict_proba", "kneighbors"],
-        "dataset": "classifier",
-    },
-    {
-        "model": KNeighborsClassifier(
-            n_neighbors=10, algorithm="kd_tree", weights="distance"
-        ),
-        "methods": ["predict", "predict_proba", "kneighbors"],
-        "dataset": "classifier",
-    },
-    {
-        "model": KNeighborsRegressor(
-            n_neighbors=10, algorithm="kd_tree", weights="distance"
-        ),
-        "methods": ["predict", "kneighbors"],
-        "dataset": "regression",
-    },
-    {
-        "model": KNeighborsRegressor(
-            n_neighbors=10, algorithm="kd_tree", weights="uniform"
-        ),
-        "methods": ["predict", "kneighbors"],
-        "dataset": "regression",
-    },
-    {
-        "model": KNeighborsRegressor(
-            n_neighbors=10, algorithm="brute", weights="distance"
-        ),
-        "methods": ["predict", "kneighbors"],
-        "dataset": "regression",
-    },
-    {
-        "model": KNeighborsRegressor(
-            n_neighbors=10, algorithm="brute", weights="uniform"
-        ),
-        "methods": ["predict", "kneighbors"],
-        "dataset": "regression",
-    },
-    {
-        "model": NearestNeighbors(n_neighbors=10, algorithm="brute"),
-        "methods": ["kneighbors"],
-        "dataset": "blobs",
-    },
-    {
-        "model": NearestNeighbors(n_neighbors=10, algorithm="kd_tree"),
-        "methods": ["kneighbors"],
-        "dataset": "blobs",
-    },
-    {
-        "model": LocalOutlierFactor(n_neighbors=10, novelty=False),
-        "methods": ["fit_predict"],
-        "dataset": "blobs",
-    },
-    {
-        "model": LocalOutlierFactor(n_neighbors=10, novelty=True),
-        "methods": ["predict"],
-        "dataset": "blobs",
-    },
-    {
-        "model": DBSCAN(algorithm="brute", n_jobs=-1),
-        "methods": [],
-        "dataset": "blobs",
-    },
-    {
-        "model": SVC(kernel="rbf"),
-        "methods": ["predict", "decision_function"],
-        "dataset": "classifier",
-    },
-    {
-        "model": SVC(kernel="rbf"),
-        "methods": ["predict", "decision_function"],
-        "dataset": "sparse",
-    },
-    {
-        "model": NuSVC(kernel="rbf"),
-        "methods": ["predict", "decision_function"],
-        "dataset": "classifier",
-    },
-    {
-        "model": SVR(kernel="rbf"),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": NuSVR(kernel="rbf"),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": TSNE(random_state=0),
-        "methods": ["fit_transform"],
-        "dataset": "classifier",
-    },
-    {
-        "model": KMeans(random_state=0, init="k-means++"),
-        "methods": ["predict"],
-        "dataset": "blobs",
-    },
-    {
-        "model": KMeans(random_state=0, init="random"),
-        "methods": ["predict"],
-        "dataset": "blobs",
-    },
-    {
-        "model": KMeans(random_state=0, init="k-means++"),
-        "methods": ["predict"],
-        "dataset": "sparse",
-    },
-    {
-        "model": KMeans(random_state=0, init="random"),
-        "methods": ["predict"],
-        "dataset": "sparse",
-    },
-    {
-        "model": ElasticNet(random_state=0),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": Lasso(random_state=0),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": PCA(n_components=0.5, svd_solver="full", random_state=0),
-        "methods": ["transform", "get_covariance", "get_precision", "score_samples"],
-        "dataset": "classifier",
-    },
-    {
-        "model": RandomForestClassifier(
-            random_state=0, oob_score=True, max_samples=0.5, max_features="sqrt"
-        ),
-        "methods": ["predict", "predict_proba"],
-        "dataset": "classifier",
-    },
-    {
-        "model": LogisticRegression(random_state=0, solver="newton-cg", max_iter=1000),
-        "methods": ["predict", "predict_proba"],
-        "dataset": "classifier",
-    },
-    {
-        "model": LogisticRegression(random_state=0, solver="lbfgs", max_iter=1000),
-        "methods": ["predict", "predict_proba"],
-        "dataset": "classifier",
-    },
-    {
-        "model": LogisticRegressionCV(
-            random_state=0, solver="newton-cg", n_jobs=-1, max_iter=1000
-        ),
-        "methods": ["predict", "predict_proba"],
-        "dataset": "classifier",
-    },
-    {
-        "model": LogisticRegressionCV(
-            random_state=0, solver="lbfgs", n_jobs=-1, max_iter=1000
-        ),
-        "methods": ["predict", "predict_proba"],
-        "dataset": "classifier",
-    },
-    {
-        "model": RandomForestRegressor(
-            random_state=0, oob_score=True, max_samples=0.5, max_features="sqrt"
-        ),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": LinearRegression(),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-    {
-        "model": Ridge(random_state=0),
-        "methods": ["predict"],
-        "dataset": "regression",
-    },
-]
-
-TO_SKIP = [
-    "TSNE",  # Absolute diff is 1e-10, potential problem in KNN,
-    # will be fixed for next release. (UPD. KNN is fixed but there is a problem
-    # with stability of stock sklearn. It is already stable in master, so, we
-    # need to wait for the next sklearn release)
-    "LogisticRegression",  # Absolute diff is 1e-8, will be fixed for next release
-    "LogisticRegressionCV",  # Absolute diff is 1e-10, will be fixed for next release
-    "RandomForestRegressor",  # Absolute diff is 1e-14 in OOB score,
-    # will be fixed for next release
-]
-
-
-@pytest.mark.parametrize("model_head", MODELS_INFO)
-def test_models(model_head):
-    stable_algos = []
-    if get_class_name(model_head["model"]) in stable_algos and daal_check_version(
-        (2021, "P", 300)
-    ):
-        try:
-            TO_SKIP.remove(get_class_name(model_head["model"]))
-        except ValueError:
-            pass
-    if get_class_name(model_head["model"]) in TO_SKIP:
-        pytest.skip("Unstable", allow_module_level=False)
-    _run_test(model_head["model"], model_head["methods"], model_head["dataset"])
-
-
-@pytest.mark.parametrize("features", range(5, 10))
-def test_train_test_split(features):
-    X, y = make_classification(
-        n_samples=4000,
-        n_features=features,
-        n_informative=features,
-        n_redundant=0,
-        n_clusters_per_class=8,
-        random_state=0,
-    )
-    (
-        baseline_X_train,
-        baseline_X_test,
-        baseline_y_train,
-        baseline_y_test,
-    ) = train_test_split(X, y, test_size=0.33, random_state=0)
-    baseline = [baseline_X_train, baseline_X_test, baseline_y_train, baseline_y_test]
-    for _ in range(10):
-        X_train, X_test, y_train, y_test = train_test_split(
-            X, y, test_size=0.33, random_state=0
-        )
-        res = [X_train, X_test, y_train, y_test]
-        for a, b in zip(res, baseline):
-            np.testing.assert_allclose(
-                a, b, rtol=0.0, atol=0.0, err_msg=str("train_test_split is incorrect")
-            )
-
-
-@pytest.mark.parametrize("metric", ["cosine", "correlation"])
-def test_pairwise_distances(metric):
-    X = np.random.rand(1000)
-    X = np.array(X, dtype=np.float64)
-    baseline = pairwise_distances(X.reshape(1, -1), metric=metric)
-    for _ in range(5):
-        res = pairwise_distances(X.reshape(1, -1), metric=metric)
-        for a, b in zip(res, baseline):
-            np.testing.assert_allclose(
-                a, b, rtol=0.0, atol=0.0, err_msg=str("pairwise_distances is incorrect")
-            )
-
-
-@pytest.mark.parametrize("array_size", [100, 1000, 10000])
-def test_roc_auc(array_size):
-    a = [random.randint(0, 1) for i in range(array_size)]
-    b = [random.randint(0, 1) for i in range(array_size)]
-    baseline = roc_auc_score(a, b)
-    for _ in range(5):
-        res = roc_auc_score(a, b)
-        np.testing.assert_allclose(
-            baseline, res, rtol=0.0, atol=0.0, err_msg=str("roc_auc is incorrect")
-        )

scikit_learn_intelex-2024.4.0.data/data/Lib/site-packages/sklearnex/utils/__init__.py

@@ -1,20 +0,0 @@
-# ===============================================================================
-# Copyright 2022 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 ._namespace import get_namespace
-from .validation import _assert_all_finite
-
-__all__ = ["get_namespace", "_assert_all_finite"]

scikit_learn_intelex-2024.4.0.data/data/Lib/site-packages/sklearnex/utils/_namespace.py

@@ -1,97 +0,0 @@
-# ==============================================================================
-# Copyright 2024 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
-
-from daal4py.sklearn._utils import sklearn_check_version
-
-from .._device_offload import dpnp_available
-
-if sklearn_check_version("1.2"):
-    from sklearn.utils._array_api import get_namespace as sklearn_get_namespace
-
-if dpnp_available:
-    import dpnp
-
-
-def get_namespace(*arrays):
-    """Get namespace of arrays.
-
-    Introspect `arrays` arguments and return their common Array API
-    compatible namespace object, if any. NumPy 1.22 and later can
-    construct such containers using the `numpy.array_api` namespace
-    for instance.
-
-    This function will return the namespace of SYCL-related arrays
-    which define the __sycl_usm_array_interface__ attribute
-    regardless of array_api support, the configuration of
-    array_api_dispatch, or scikit-learn version.
-
-    See: https://numpy.org/neps/nep-0047-array-api-standard.html
-
-    If `arrays` are regular numpy arrays, an instance of the
-    `_NumPyApiWrapper` compatibility wrapper is returned instead.
-
-    Namespace support is not enabled by default. To enabled it
-    call:
-
-      sklearn.set_config(array_api_dispatch=True)
-
-    or:
-
-      with sklearn.config_context(array_api_dispatch=True):
-          # your code here
-
-    Otherwise an instance of the `_NumPyApiWrapper`
-    compatibility wrapper is always returned irrespective of
-    the fact that arrays implement the `__array_namespace__`
-    protocol or not.
-
-    Parameters
-    ----------
-    *arrays : array objects
-        Array objects.
-
-    Returns
-    -------
-    namespace : module
-        Namespace shared by array objects.
-
-    is_array_api : bool
-        True of the arrays are containers that implement the Array API spec.
-    """
-
-    # sycl support designed to work regardless of array_api_dispatch sklearn global value
-    sycl_type = {type(x): x for x in arrays if hasattr(x, "__sycl_usm_array_interface__")}
-
-    if len(sycl_type) > 1:
-        raise ValueError(f"Multiple SYCL types for array inputs: {sycl_type}")
-
-    if sycl_type:
-
-        (X,) = sycl_type.values()
-
-        if hasattr(X, "__array_namespace__"):
-            return X.__array_namespace__(), True
-        elif dpnp_available and isinstance(X, dpnp.ndarray):
-            return dpnp, False
-        else:
-            raise ValueError(f"SYCL type not recognized: {sycl_type}")
-
-    elif sklearn_check_version("1.2"):
-        return sklearn_get_namespace(*arrays)
-    else:
-        return np, True