scikit-learn-intelex 2024.1.0__py312-none-win_amd64.whl → 2024.3.0__py312-none-win_amd64.whl

{scikit_learn_intelex-2024.1.0.data → scikit_learn_intelex-2024.3.0.data}/data/Lib/site-packages/sklearnex/__init__.py

@@ -1,5 +1,6 @@
 # ==============================================================================
 # Copyright 2021 Intel Corporation
+# Copyright 2024 Fujitsu Limited
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -14,7 +15,7 @@
 # limitations under the License.
 # ==============================================================================
 
-from onedal.common.hyperparameters import get_hyperparameters
+import os
 
 from . import utils
 from ._config import config_context, get_config, set_config
@@ -41,21 +42,22 @@ __all__ = [
     "linear_model",
     "manifold",
     "metrics",
+    "model_selection",
     "neighbors",
     "patch_sklearn",
     "set_config",
     "sklearn_is_patched",
-    "sklearn_is_patchedget_patch_map",
     "svm",
     "unpatch_sklearn",
     "utils",
 ]
+onedal_iface_flag = os.environ.get("OFF_ONEDAL_IFACE", "0")
+if onedal_iface_flag == "0":
+    from onedal import _is_spmd_backend
+    from onedal.common.hyperparameters import get_hyperparameters
 
-
-from onedal import _is_dpc_backend
-
-if _is_dpc_backend:
-    __all__.append("spmd")
+    if _is_spmd_backend:
+        __all__.append("spmd")
 
 
 from ._utils import set_sklearn_ex_verbose

{scikit_learn_intelex-2024.1.0.data → scikit_learn_intelex-2024.3.0.data}/data/Lib/site-packages/sklearnex/cluster/dbscan.py

@@ -22,10 +22,11 @@ from scipy import sparse as sp
 from sklearn.cluster import DBSCAN as sklearn_DBSCAN
 from sklearn.utils.validation import _check_sample_weight
 
-from daal4py.sklearn._utils import control_n_jobs, run_with_n_jobs, sklearn_check_version
+from daal4py.sklearn._n_jobs_support import control_n_jobs
+from daal4py.sklearn._utils import sklearn_check_version
 from onedal.cluster import DBSCAN as onedal_DBSCAN
 
-from .._device_offload import dispatch, wrap_output_data
+from .._device_offload import dispatch
 from .._utils import PatchingConditionsChain
 
 if sklearn_check_version("1.1") and not sklearn_check_version("1.2"):
@@ -45,7 +46,7 @@ class BaseDBSCAN(ABC):
         self.n_features_in_ = self._onedal_estimator.n_features_in_
 
 
-@control_n_jobs
+@control_n_jobs(decorated_methods=["fit"])
 class DBSCAN(sklearn_DBSCAN, BaseDBSCAN):
     __doc__ = sklearn_DBSCAN.__doc__
 
@@ -83,7 +84,6 @@ class DBSCAN(sklearn_DBSCAN, BaseDBSCAN):
         self.p = p
         self.n_jobs = n_jobs
 
-    @run_with_n_jobs
     def _onedal_fit(self, X, y, sample_weight=None, queue=None):
         onedal_params = {
             "eps": self.eps,
@@ -186,3 +186,5 @@ class DBSCAN(sklearn_DBSCAN, BaseDBSCAN):
         )
 
         return self
+
+    fit.__doc__ = sklearn_DBSCAN.fit.__doc__

scikit_learn_intelex-2024.3.0.data/data/Lib/site-packages/sklearnex/conftest.py

@@ -0,0 +1,63 @@
+# ==============================================================================
+# 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 io
+import logging
+
+import pytest
+
+from sklearnex import patch_sklearn, unpatch_sklearn
+
+
+def pytest_configure(config):
+    config.addinivalue_line(
+        "markers", "allow_sklearn_fallback: mark test to not check for sklearnex usage"
+    )
+
+
+@pytest.hookimpl(hookwrapper=True)
+def pytest_runtest_call(item):
+    # setup logger to check for sklearn fallback
+    if not item.get_closest_marker("allow_sklearn_fallback"):
+        log_stream = io.StringIO()
+        log_handler = logging.StreamHandler(log_stream)
+        sklearnex_logger = logging.getLogger("sklearnex")
+        level = sklearnex_logger.level
+        sklearnex_stderr_handler = sklearnex_logger.handlers
+        sklearnex_logger.handlers = []
+        sklearnex_logger.addHandler(log_handler)
+        sklearnex_logger.setLevel(logging.INFO)
+        log_handler.setLevel(logging.INFO)
+
+        yield
+
+        sklearnex_logger.handlers = sklearnex_stderr_handler
+        sklearnex_logger.setLevel(level)
+        sklearnex_logger.removeHandler(log_handler)
+        text = log_stream.getvalue()
+        if "fallback to original Scikit-learn" in text:
+            raise TypeError(
+                f"test did not properly evaluate sklearnex functionality and fell back to sklearn:\n{text}"
+            )
+    else:
+        yield
+
+
+@pytest.fixture
+def with_sklearnex():
+    patch_sklearn()
+    yield
+    unpatch_sklearn()

scikit_learn_intelex-2024.3.0.data/data/Lib/site-packages/sklearnex/covariance/__init__.py

@@ -0,0 +1,19 @@
+# ===============================================================================
+# 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.
+# ===============================================================================
+
+from .incremental_covariance import IncrementalEmpiricalCovariance
+
+__all__ = ["IncrementalEmpiricalCovariance"]

scikit_learn_intelex-2024.3.0.data/data/Lib/site-packages/sklearnex/covariance/incremental_covariance.py

@@ -0,0 +1,130 @@
+# ===============================================================================
+# 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 sklearn.utils import check_array, gen_batches
+
+from daal4py.sklearn._n_jobs_support import control_n_jobs
+from onedal._device_offload import support_usm_ndarray
+from onedal.covariance import (
+    IncrementalEmpiricalCovariance as onedal_IncrementalEmpiricalCovariance,
+)
+
+
+@control_n_jobs(decorated_methods=["partial_fit"])
+class IncrementalEmpiricalCovariance:
+    """
+    Incremental estimator for covariance.
+    Allows to compute empirical covariance estimated by maximum
+    likelihood method if data are splitted into batches.
+
+    Parameters
+    ----------
+    batch_size : int, default=None
+        The number of samples to use for each batch. Only used when calling
+        ``fit``. If ``batch_size`` is ``None``, then ``batch_size``
+        is inferred from the data and set to ``5 * n_features``, to provide a
+        balance between approximation accuracy and memory consumption.
+
+    Attributes
+    ----------
+    location_ : ndarray of shape (n_features,)
+        Estimated location, i.e. the estimated mean.
+
+    covariance_ : ndarray of shape (n_features, n_features)
+        Estimated covariance matrix
+    """
+
+    _onedal_incremental_covariance = staticmethod(onedal_IncrementalEmpiricalCovariance)
+
+    def __init__(self, batch_size=None):
+        self._need_to_finalize = False  # If True then finalize compute should
+        #      be called to obtain covariance_ or location_ from partial compute data
+        self.batch_size = batch_size
+
+    def _onedal_finalize_fit(self):
+        assert hasattr(self, "_onedal_estimator")
+        self._onedal_estimator.finalize_fit()
+        self._need_to_finalize = False
+
+    def _onedal_partial_fit(self, X, queue):
+        onedal_params = {
+            "method": "dense",
+            "bias": True,
+        }
+        if not hasattr(self, "_onedal_estimator"):
+            self._onedal_estimator = self._onedal_incremental_covariance(**onedal_params)
+        self._onedal_estimator.partial_fit(X, queue)
+        self._need_to_finalize = True
+
+    @property
+    def covariance_(self):
+        if self._need_to_finalize:
+            self._onedal_finalize_fit()
+        return self._onedal_estimator.covariance_
+
+    @property
+    def location_(self):
+        if self._need_to_finalize:
+            self._onedal_finalize_fit()
+        return self._onedal_estimator.location_
+
+    @support_usm_ndarray()
+    def partial_fit(self, X, queue=None):
+        """
+        Incremental fit with X. All of X is processed as a single batch.
+
+        Parameters
+        ----------
+        X : array-like of shape (n_samples, n_features)
+            Training data, where `n_samples` is the number of samples and
+            `n_features` is the number of features.
+
+        Returns
+        -------
+        self : object
+            Returns the instance itself.
+        """
+        X = check_array(X, dtype=[np.float64, np.float32])
+        self._onedal_partial_fit(X, queue)
+        return self
+
+    def fit(self, X, queue=None):
+        """
+        Fit the model with X, using minibatches of size batch_size.
+
+        Parameters
+        ----------
+        X : array-like of shape (n_samples, n_features)
+            Training data, where `n_samples` is the number of samples and
+            `n_features` is the number of features.
+
+        Returns
+        -------
+        self : object
+            Returns the instance itself.
+        """
+        n_samples, n_features = X.shape
+        if self.batch_size is None:
+            batch_size_ = 5 * n_features
+        else:
+            batch_size_ = self.batch_size
+        for batch in gen_batches(n_samples, batch_size_):
+            X_batch = X[batch]
+            self.partial_fit(X_batch, queue=queue)
+
+        self._onedal_finalize_fit()
+        return self

scikit_learn_intelex-2024.3.0.data/data/Lib/site-packages/sklearnex/covariance/tests/test_incremental_covariance.py

@@ -0,0 +1,143 @@
+# ===============================================================================
+# 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
+import pytest
+from numpy.testing import assert_allclose
+
+from onedal.tests.utils._dataframes_support import (
+    _convert_to_dataframe,
+    get_dataframes_and_queues,
+)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_sklearnex_partial_fit_on_gold_data(dataframe, queue, dtype):
+    from sklearnex.covariance import IncrementalEmpiricalCovariance
+
+    X = np.array([[0, 1], [0, 1]])
+    X = X.astype(dtype)
+    X_split = np.array_split(X, 2)
+    inccov = IncrementalEmpiricalCovariance()
+
+    for i in range(2):
+        X_split_df = _convert_to_dataframe(
+            X_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        result = inccov.partial_fit(X_split_df)
+
+    expected_covariance = np.array([[0, 0], [0, 0]])
+    expected_means = np.array([0, 1])
+
+    assert_allclose(expected_covariance, result.covariance_)
+    assert_allclose(expected_means, result.location_)
+
+    X = np.array([[1, 2], [3, 6]])
+    X = X.astype(dtype)
+    X_split = np.array_split(X, 2)
+    inccov = IncrementalEmpiricalCovariance()
+
+    for i in range(2):
+        X_split_df = _convert_to_dataframe(
+            X_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        result = inccov.partial_fit(X_split_df)
+
+    expected_covariance = np.array([[1, 2], [2, 4]])
+    expected_means = np.array([2, 4])
+
+    assert_allclose(expected_covariance, result.covariance_)
+    assert_allclose(expected_means, result.location_)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("batch_size", [2, 4])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_sklearnex_fit_on_gold_data(dataframe, queue, batch_size, dtype):
+    from sklearnex.covariance import IncrementalEmpiricalCovariance
+
+    X = np.array([[0, 1, 2, 3], [0, -1, -2, -3], [0, 1, 2, 3], [0, 1, 2, 3]])
+    X = X.astype(dtype)
+    X_df = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    inccov = IncrementalEmpiricalCovariance(batch_size=batch_size)
+
+    result = inccov.fit(X_df)
+
+    expected_covariance = np.array(
+        [[0, 0, 0, 0], [0, 0.75, 1.5, 2.25], [0, 1.5, 3, 4.5], [0, 2.25, 4.5, 6.75]]
+    )
+    expected_means = np.array([0, 0.5, 1, 1.5])
+
+    assert_allclose(expected_covariance, result.covariance_)
+    assert_allclose(expected_means, result.location_)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("num_batches", [2, 4, 6, 8, 10])
+@pytest.mark.parametrize("row_count", [100, 1000, 2000])
+@pytest.mark.parametrize("column_count", [10, 100, 200])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_sklearnex_partial_fit_on_random_data(
+    dataframe, queue, num_batches, row_count, column_count, dtype
+):
+    from sklearnex.covariance import IncrementalEmpiricalCovariance
+
+    seed = 77
+    gen = np.random.default_rng(seed)
+    X = gen.uniform(low=-0.3, high=+0.7, size=(row_count, column_count))
+    X = X.astype(dtype)
+    X_split = np.array_split(X, num_batches)
+    inccov = IncrementalEmpiricalCovariance()
+
+    for i in range(num_batches):
+        X_split_df = _convert_to_dataframe(
+            X_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        result = inccov.partial_fit(X_split_df)
+
+    expected_covariance = np.cov(X.T, bias=1)
+    expected_means = np.mean(X, axis=0)
+
+    assert_allclose(expected_covariance, result.covariance_, atol=1e-6)
+    assert_allclose(expected_means, result.location_, atol=1e-6)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("num_batches", [2, 4, 6, 8, 10])
+@pytest.mark.parametrize("row_count", [100, 1000, 2000])
+@pytest.mark.parametrize("column_count", [10, 100, 200])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_sklearnex_fit_on_random_data(
+    dataframe, queue, num_batches, row_count, column_count, dtype
+):
+    from sklearnex.covariance import IncrementalEmpiricalCovariance
+
+    seed = 77
+    gen = np.random.default_rng(seed)
+    X = gen.uniform(low=-0.3, high=+0.7, size=(row_count, column_count))
+    X = X.astype(dtype)
+    X_df = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    batch_size = row_count // num_batches
+    inccov = IncrementalEmpiricalCovariance(batch_size=batch_size)
+
+    result = inccov.fit(X_df)
+
+    expected_covariance = np.cov(X.T, bias=1)
+    expected_means = np.mean(X, axis=0)
+
+    assert_allclose(expected_covariance, result.covariance_, atol=1e-6)
+    assert_allclose(expected_means, result.location_, atol=1e-6)