scikit-learn-intelex 2024.3.0__py311-none-win_amd64.whl → 2024.5.0__py311-none-win_amd64.whl

scikit_learn_intelex-2024.5.0.data/data/Lib/site-packages/sklearnex/basic_statistics/tests/test_incremental_basic_statistics.py

@@ -0,0 +1,384 @@
+# ===============================================================================
+# 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.basic_statistics.tests.test_incremental_basic_statistics import (
+    expected_max,
+    expected_mean,
+    expected_sum,
+    options_and_tests,
+)
+from onedal.tests.utils._dataframes_support import (
+    _convert_to_dataframe,
+    get_dataframes_and_queues,
+)
+from sklearnex.basic_statistics import IncrementalBasicStatistics
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("weighted", [True, False])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_partial_fit_multiple_options_on_gold_data(dataframe, queue, weighted, dtype):
+    X = np.array([[0, 0], [1, 1]])
+    X = X.astype(dtype=dtype)
+    X_split = np.array_split(X, 2)
+    if weighted:
+        weights = np.array([1, 0.5])
+        weights = weights.astype(dtype=dtype)
+        weights_split = np.array_split(weights, 2)
+
+    incbs = IncrementalBasicStatistics()
+    for i in range(2):
+        X_split_df = _convert_to_dataframe(
+            X_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        if weighted:
+            weights_split_df = _convert_to_dataframe(
+                weights_split[i], sycl_queue=queue, target_df=dataframe
+            )
+            result = incbs.partial_fit(X_split_df, sample_weight=weights_split_df)
+        else:
+            result = incbs.partial_fit(X_split_df)
+
+    if weighted:
+        expected_weighted_mean = np.array([0.25, 0.25])
+        expected_weighted_min = np.array([0, 0])
+        expected_weighted_max = np.array([0.5, 0.5])
+        assert_allclose(expected_weighted_mean, result.mean)
+        assert_allclose(expected_weighted_max, result.max)
+        assert_allclose(expected_weighted_min, result.min)
+    else:
+        expected_mean = np.array([0.5, 0.5])
+        expected_min = np.array([0, 0])
+        expected_max = np.array([1, 1])
+        assert_allclose(expected_mean, result.mean)
+        assert_allclose(expected_max, result.max)
+        assert_allclose(expected_min, result.min)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("num_batches", [2, 10])
+@pytest.mark.parametrize("option", options_and_tests)
+@pytest.mark.parametrize("row_count", [100, 1000])
+@pytest.mark.parametrize("column_count", [10, 100])
+@pytest.mark.parametrize("weighted", [True, False])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_partial_fit_single_option_on_random_data(
+    dataframe, queue, num_batches, option, row_count, column_count, weighted, dtype
+):
+    result_option, function, tols = option
+    fp32tol, fp64tol = tols
+    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=dtype)
+    X_split = np.array_split(X, num_batches)
+    if weighted:
+        weights = gen.uniform(low=-0.5, high=+1.0, size=row_count)
+        weights = weights.astype(dtype=dtype)
+        weights_split = np.array_split(weights, num_batches)
+    incbs = IncrementalBasicStatistics(result_options=result_option)
+
+    for i in range(num_batches):
+        X_split_df = _convert_to_dataframe(
+            X_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        if weighted:
+            weights_split_df = _convert_to_dataframe(
+                weights_split[i], sycl_queue=queue, target_df=dataframe
+            )
+            result = incbs.partial_fit(X_split_df, sample_weight=weights_split_df)
+        else:
+            result = incbs.partial_fit(X_split_df)
+
+    res = getattr(result, result_option)
+    if weighted:
+        weighted_data = np.diag(weights) @ X
+        gtr = function(weighted_data)
+    else:
+        gtr = function(X)
+
+    tol = fp32tol if res.dtype == np.float32 else fp64tol
+    assert_allclose(gtr, res, atol=tol)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("num_batches", [2, 10])
+@pytest.mark.parametrize("row_count", [100, 1000])
+@pytest.mark.parametrize("column_count", [10, 100])
+@pytest.mark.parametrize("weighted", [True, False])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_partial_fit_multiple_options_on_random_data(
+    dataframe, queue, num_batches, row_count, column_count, weighted, dtype
+):
+    seed = 42
+    gen = np.random.default_rng(seed)
+    X = gen.uniform(low=-0.3, high=+0.7, size=(row_count, column_count))
+    X = X.astype(dtype=dtype)
+    X_split = np.array_split(X, num_batches)
+    if weighted:
+        weights = gen.uniform(low=-0.5, high=+1.0, size=row_count)
+        weights = weights.astype(dtype=dtype)
+        weights_split = np.array_split(weights, num_batches)
+    incbs = IncrementalBasicStatistics(result_options=["mean", "max", "sum"])
+
+    for i in range(num_batches):
+        X_split_df = _convert_to_dataframe(
+            X_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        if weighted:
+            weights_split_df = _convert_to_dataframe(
+                weights_split[i], sycl_queue=queue, target_df=dataframe
+            )
+            result = incbs.partial_fit(X_split_df, sample_weight=weights_split_df)
+        else:
+            result = incbs.partial_fit(X_split_df)
+
+    res_mean, res_max, res_sum = result.mean, result.max, result.sum
+    if weighted:
+        weighted_data = np.diag(weights) @ X
+        gtr_mean, gtr_max, gtr_sum = (
+            expected_mean(weighted_data),
+            expected_max(weighted_data),
+            expected_sum(weighted_data),
+        )
+    else:
+        gtr_mean, gtr_max, gtr_sum = (
+            expected_mean(X),
+            expected_max(X),
+            expected_sum(X),
+        )
+
+    tol = 3e-4 if res_mean.dtype == np.float32 else 1e-7
+    assert_allclose(gtr_mean, res_mean, atol=tol)
+    assert_allclose(gtr_max, res_max, atol=tol)
+    assert_allclose(gtr_sum, res_sum, atol=tol)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("num_batches", [2, 10])
+@pytest.mark.parametrize("row_count", [100, 1000])
+@pytest.mark.parametrize("column_count", [10, 100])
+@pytest.mark.parametrize("weighted", [True, False])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_partial_fit_all_option_on_random_data(
+    dataframe, queue, num_batches, row_count, column_count, weighted, dtype
+):
+    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=dtype)
+    X_split = np.array_split(X, num_batches)
+    if weighted:
+        weights = gen.uniform(low=-0.5, high=+1.0, size=row_count)
+        weights = weights.astype(dtype=dtype)
+        weights_split = np.array_split(weights, num_batches)
+    incbs = IncrementalBasicStatistics(result_options="all")
+
+    for i in range(num_batches):
+        X_split_df = _convert_to_dataframe(
+            X_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        if weighted:
+            weights_split_df = _convert_to_dataframe(
+                weights_split[i], sycl_queue=queue, target_df=dataframe
+            )
+            result = incbs.partial_fit(X_split_df, sample_weight=weights_split_df)
+        else:
+            result = incbs.partial_fit(X_split_df)
+
+    if weighted:
+        weighted_data = np.diag(weights) @ X
+
+    for option in options_and_tests:
+        result_option, function, tols = option
+        fp32tol, fp64tol = tols
+        res = getattr(result, result_option)
+        if weighted:
+            gtr = function(weighted_data)
+        else:
+            gtr = function(X)
+        tol = fp32tol if res.dtype == np.float32 else fp64tol
+        assert_allclose(gtr, res, atol=tol)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("weighted", [True, False])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_fit_multiple_options_on_gold_data(dataframe, queue, weighted, dtype):
+    X = np.array([[0, 0], [1, 1]])
+    X = X.astype(dtype=dtype)
+    X_df = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    if weighted:
+        weights = np.array([1, 0.5])
+        weights = weights.astype(dtype=dtype)
+        weights_df = _convert_to_dataframe(weights, sycl_queue=queue, target_df=dataframe)
+    incbs = IncrementalBasicStatistics(batch_size=1)
+
+    if weighted:
+        result = incbs.fit(X_df, sample_weight=weights_df)
+    else:
+        result = incbs.fit(X_df)
+
+    if weighted:
+        expected_weighted_mean = np.array([0.25, 0.25])
+        expected_weighted_min = np.array([0, 0])
+        expected_weighted_max = np.array([0.5, 0.5])
+        assert_allclose(expected_weighted_mean, result.mean)
+        assert_allclose(expected_weighted_max, result.max)
+        assert_allclose(expected_weighted_min, result.min)
+    else:
+        expected_mean = np.array([0.5, 0.5])
+        expected_min = np.array([0, 0])
+        expected_max = np.array([1, 1])
+        assert_allclose(expected_mean, result.mean)
+        assert_allclose(expected_max, result.max)
+        assert_allclose(expected_min, result.min)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("num_batches", [2, 10])
+@pytest.mark.parametrize("option", options_and_tests)
+@pytest.mark.parametrize("row_count", [100, 1000])
+@pytest.mark.parametrize("column_count", [10, 100])
+@pytest.mark.parametrize("weighted", [True, False])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_fit_single_option_on_random_data(
+    dataframe, queue, num_batches, option, row_count, column_count, weighted, dtype
+):
+    result_option, function, tols = option
+    fp32tol, fp64tol = tols
+    seed = 77
+    gen = np.random.default_rng(seed)
+    batch_size = row_count // num_batches
+    X = gen.uniform(low=-0.3, high=+0.7, size=(row_count, column_count))
+    X = X.astype(dtype=dtype)
+    X_df = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    if weighted:
+        weights = gen.uniform(low=-0.5, high=1.0, size=row_count)
+        weights = weights.astype(dtype=dtype)
+        weights_df = _convert_to_dataframe(weights, sycl_queue=queue, target_df=dataframe)
+    incbs = IncrementalBasicStatistics(
+        result_options=result_option, batch_size=batch_size
+    )
+
+    if weighted:
+        result = incbs.fit(X_df, sample_weight=weights_df)
+    else:
+        result = incbs.fit(X_df)
+
+    res = getattr(result, result_option)
+    if weighted:
+        weighted_data = np.diag(weights) @ X
+        gtr = function(weighted_data)
+    else:
+        gtr = function(X)
+
+    tol = fp32tol if res.dtype == np.float32 else fp64tol
+    assert_allclose(gtr, res, atol=tol)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("num_batches", [2, 10])
+@pytest.mark.parametrize("row_count", [100, 1000])
+@pytest.mark.parametrize("column_count", [10, 100])
+@pytest.mark.parametrize("weighted", [True, False])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_fit_multiple_options_on_random_data(
+    dataframe, queue, num_batches, row_count, column_count, weighted, dtype
+):
+    seed = 77
+    gen = np.random.default_rng(seed)
+    batch_size = row_count // num_batches
+    X = gen.uniform(low=-0.3, high=+0.7, size=(row_count, column_count))
+    X = X.astype(dtype=dtype)
+    X_df = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    if weighted:
+        weights = gen.uniform(low=-0.5, high=1.0, size=row_count)
+        weights = weights.astype(dtype=dtype)
+        weights_df = _convert_to_dataframe(weights, sycl_queue=queue, target_df=dataframe)
+    incbs = IncrementalBasicStatistics(
+        result_options=["mean", "max", "sum"], batch_size=batch_size
+    )
+
+    if weighted:
+        result = incbs.fit(X_df, sample_weight=weights_df)
+    else:
+        result = incbs.fit(X_df)
+
+    res_mean, res_max, res_sum = result.mean, result.max, result.sum
+    if weighted:
+        weighted_data = np.diag(weights) @ X
+        gtr_mean, gtr_max, gtr_sum = (
+            expected_mean(weighted_data),
+            expected_max(weighted_data),
+            expected_sum(weighted_data),
+        )
+    else:
+        gtr_mean, gtr_max, gtr_sum = (
+            expected_mean(X),
+            expected_max(X),
+            expected_sum(X),
+        )
+
+    tol = 3e-4 if res_mean.dtype == np.float32 else 1e-7
+    assert_allclose(gtr_mean, res_mean, atol=tol)
+    assert_allclose(gtr_max, res_max, atol=tol)
+    assert_allclose(gtr_sum, res_sum, atol=tol)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("num_batches", [2, 10])
+@pytest.mark.parametrize("row_count", [100, 1000])
+@pytest.mark.parametrize("column_count", [10, 100])
+@pytest.mark.parametrize("weighted", [True, False])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_fit_all_option_on_random_data(
+    dataframe, queue, num_batches, row_count, column_count, weighted, dtype
+):
+    seed = 77
+    gen = np.random.default_rng(seed)
+    batch_size = row_count // num_batches
+    X = gen.uniform(low=-0.3, high=+0.7, size=(row_count, column_count))
+    X = X.astype(dtype=dtype)
+    X_df = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    if weighted:
+        weights = gen.uniform(low=-0.5, high=+1.0, size=row_count)
+        weights = weights.astype(dtype=dtype)
+        weights_df = _convert_to_dataframe(weights, sycl_queue=queue, target_df=dataframe)
+    incbs = IncrementalBasicStatistics(result_options="all", batch_size=batch_size)
+
+    if weighted:
+        result = incbs.fit(X_df, sample_weight=weights_df)
+    else:
+        result = incbs.fit(X_df)
+
+    if weighted:
+        weighted_data = np.diag(weights) @ X
+
+    for option in options_and_tests:
+        result_option, function, tols = option
+        fp32tol, fp64tol = tols
+        res = getattr(result, result_option)
+        if weighted:
+            gtr = function(weighted_data)
+        else:
+            gtr = function(X)
+        tol = fp32tol if res.dtype == np.float32 else fp64tol
+        assert_allclose(gtr, res, atol=tol)

scikit_learn_intelex-2024.5.0.data/data/Lib/site-packages/sklearnex/covariance/incremental_covariance.py

@@ -0,0 +1,317 @@
+# ===============================================================================
+# 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 numbers
+import warnings
+
+import numpy as np
+from scipy import linalg
+from sklearn.base import BaseEstimator
+from sklearn.covariance import EmpiricalCovariance as sklearn_EmpiricalCovariance
+from sklearn.utils import check_array, gen_batches
+
+from daal4py.sklearn._n_jobs_support import control_n_jobs
+from daal4py.sklearn._utils import daal_check_version, sklearn_check_version
+from onedal._device_offload import support_usm_ndarray
+from onedal.covariance import (
+    IncrementalEmpiricalCovariance as onedal_IncrementalEmpiricalCovariance,
+)
+from sklearnex import config_context
+
+from .._device_offload import dispatch, wrap_output_data
+from .._utils import PatchingConditionsChain, register_hyperparameters
+from ..metrics import pairwise_distances
+
+if sklearn_check_version("1.2"):
+    from sklearn.utils._param_validation import Interval
+
+
+@control_n_jobs(decorated_methods=["partial_fit", "fit", "_onedal_finalize_fit"])
+class IncrementalEmpiricalCovariance(BaseEstimator):
+    """
+    Incremental estimator for covariance.
+    Allows to compute empirical covariance estimated by maximum
+    likelihood method if data are splitted into batches.
+
+    Parameters
+    ----------
+    store_precision : bool, default=False
+        Specifies if the estimated precision is stored.
+
+    assume_centered : bool, default=False
+        If True, data are not centered before computation.
+        Useful when working with data whose mean is almost, but not exactly
+        zero.
+        If False (default), data are centered before computation.
+
+    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.
+
+    copy : bool, default=True
+        If False, X will be overwritten. ``copy=False`` can be used to
+        save memory but is unsafe for general use.
+
+    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
+
+    n_samples_seen_ : int
+        The number of samples processed by the estimator. Will be reset on
+        new calls to fit, but increments across ``partial_fit`` calls.
+
+    batch_size_ : int
+        Inferred batch size from ``batch_size``.
+
+    n_features_in_ : int
+        Number of features seen during :term:`fit` `partial_fit`.
+    """
+
+    _onedal_incremental_covariance = staticmethod(onedal_IncrementalEmpiricalCovariance)
+
+    if sklearn_check_version("1.2"):
+        _parameter_constraints: dict = {
+            "store_precision": ["boolean"],
+            "assume_centered": ["boolean"],
+            "batch_size": [Interval(numbers.Integral, 1, None, closed="left"), None],
+            "copy": ["boolean"],
+        }
+
+    get_precision = sklearn_EmpiricalCovariance.get_precision
+    error_norm = wrap_output_data(sklearn_EmpiricalCovariance.error_norm)
+    score = wrap_output_data(sklearn_EmpiricalCovariance.score)
+
+    def __init__(
+        self, *, store_precision=False, assume_centered=False, batch_size=None, copy=True
+    ):
+        self.assume_centered = assume_centered
+        self.store_precision = store_precision
+        self.batch_size = batch_size
+        self.copy = copy
+
+    def _onedal_supported(self, method_name, *data):
+        patching_status = PatchingConditionsChain(
+            f"sklearn.covariance.{self.__class__.__name__}.{method_name}"
+        )
+        return patching_status
+
+    def _onedal_finalize_fit(self):
+        assert hasattr(self, "_onedal_estimator")
+        self._onedal_estimator.finalize_fit()
+        self._need_to_finalize = False
+
+        if not daal_check_version((2024, "P", 400)) and self.assume_centered:
+            location = self._onedal_estimator.location_[None, :]
+            self._onedal_estimator.covariance_ += np.dot(location.T, location)
+            self._onedal_estimator.location_ = np.zeros_like(np.squeeze(location))
+        if self.store_precision:
+            self.precision_ = linalg.pinvh(
+                self._onedal_estimator.covariance_, check_finite=False
+            )
+        else:
+            self.precision_ = None
+
+    @property
+    def covariance_(self):
+        if hasattr(self, "_onedal_estimator"):
+            if self._need_to_finalize:
+                self._onedal_finalize_fit()
+            return self._onedal_estimator.covariance_
+        else:
+            raise AttributeError(
+                f"'{self.__class__.__name__}' object has no attribute 'covariance_'"
+            )
+
+    @property
+    def location_(self):
+        if hasattr(self, "_onedal_estimator"):
+            if self._need_to_finalize:
+                self._onedal_finalize_fit()
+            return self._onedal_estimator.location_
+        else:
+            raise AttributeError(
+                f"'{self.__class__.__name__}' object has no attribute 'location_'"
+            )
+
+    def _onedal_partial_fit(self, X, queue=None, check_input=True):
+
+        first_pass = not hasattr(self, "n_samples_seen_") or self.n_samples_seen_ == 0
+
+        # finite check occurs on onedal side
+        if check_input:
+            if sklearn_check_version("1.2"):
+                self._validate_params()
+
+            if sklearn_check_version("1.0"):
+                X = self._validate_data(
+                    X,
+                    dtype=[np.float64, np.float32],
+                    reset=first_pass,
+                    copy=self.copy,
+                    force_all_finite=False,
+                )
+            else:
+                X = check_array(
+                    X,
+                    dtype=[np.float64, np.float32],
+                    copy=self.copy,
+                    force_all_finite=False,
+                )
+
+        onedal_params = {
+            "method": "dense",
+            "bias": True,
+            "assume_centered": self.assume_centered,
+        }
+        if not hasattr(self, "_onedal_estimator"):
+            self._onedal_estimator = self._onedal_incremental_covariance(**onedal_params)
+        try:
+            if first_pass:
+                self.n_samples_seen_ = X.shape[0]
+                self.n_features_in_ = X.shape[1]
+            else:
+                self.n_samples_seen_ += X.shape[0]
+
+            self._onedal_estimator.partial_fit(X, queue)
+        finally:
+            self._need_to_finalize = True
+
+        return self
+
+    def partial_fit(self, X, y=None, check_input=True):
+        """
+        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.
+
+        y : Ignored
+            Not used, present for API consistency by convention.
+
+        check_input : bool, default=True
+            Run check_array on X.
+
+        Returns
+        -------
+        self : object
+            Returns the instance itself.
+        """
+        return dispatch(
+            self,
+            "partial_fit",
+            {
+                "onedal": self.__class__._onedal_partial_fit,
+                "sklearn": None,
+            },
+            X,
+            check_input=check_input,
+        )
+
+    def fit(self, X, y=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.
+
+        y : Ignored
+            Not used, present for API consistency by convention.
+
+        Returns
+        -------
+        self : object
+            Returns the instance itself.
+        """
+
+        return dispatch(
+            self,
+            "fit",
+            {
+                "onedal": self.__class__._onedal_fit,
+                "sklearn": None,
+            },
+            X,
+        )
+
+    def _onedal_fit(self, X, queue=None):
+        self.n_samples_seen_ = 0
+        if hasattr(self, "_onedal_estimator"):
+            self._onedal_estimator._reset()
+
+        if sklearn_check_version("1.2"):
+            self._validate_params()
+
+        # finite check occurs on onedal side
+        if sklearn_check_version("1.0"):
+            X = self._validate_data(
+                X, dtype=[np.float64, np.float32], copy=self.copy, force_all_finite=False
+            )
+        else:
+            X = check_array(
+                X, dtype=[np.float64, np.float32], copy=self.copy, force_all_finite=False
+            )
+            self.n_features_in_ = X.shape[1]
+
+        self.batch_size_ = self.batch_size if self.batch_size else 5 * self.n_features_in_
+
+        if X.shape[0] == 1:
+            warnings.warn(
+                "Only one sample available. You may want to reshape your data array"
+            )
+
+        for batch in gen_batches(X.shape[0], self.batch_size_):
+            X_batch = X[batch]
+            self._onedal_partial_fit(X_batch, queue=queue, check_input=False)
+
+        self._onedal_finalize_fit()
+
+        return self
+
+    # expose sklearnex pairwise_distances if mahalanobis distance eventually supported
+    @wrap_output_data
+    def mahalanobis(self, X):
+        if sklearn_check_version("1.0"):
+            self._validate_data(X, reset=False, copy=self.copy)
+        else:
+            check_array(X, copy=self.copy)
+
+        precision = self.get_precision()
+        with config_context(assume_finite=True):
+            # compute mahalanobis distances
+            dist = pairwise_distances(
+                X, self.location_[np.newaxis, :], metric="mahalanobis", VI=precision
+            )
+
+        return np.reshape(dist, (len(X),)) ** 2
+
+    _onedal_cpu_supported = _onedal_supported
+    _onedal_gpu_supported = _onedal_supported
+
+    mahalanobis.__doc__ = sklearn_EmpiricalCovariance.mahalanobis.__doc__
+    error_norm.__doc__ = sklearn_EmpiricalCovariance.error_norm.__doc__
+    score.__doc__ = sklearn_EmpiricalCovariance.score.__doc__