scikit-learn-intelex 2024.4.0__py310-none-win_amd64.whl → 2024.6.0__py310-none-win_amd64.whl

scikit_learn_intelex-2024.6.0.data/data/Lib/site-packages/sklearnex/linear_model/tests/test_incremental_linear.py

@@ -0,0 +1,200 @@
+# ===============================================================================
+# 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 (
+    _as_numpy,
+    _convert_to_dataframe,
+    get_dataframes_and_queues,
+)
+from sklearnex.linear_model import IncrementalLinearRegression
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("fit_intercept", [True, False])
+@pytest.mark.parametrize("macro_block", [None, 1024])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_sklearnex_fit_on_gold_data(dataframe, queue, fit_intercept, macro_block, dtype):
+    X = np.array([[1], [2]])
+    X = X.astype(dtype=dtype)
+    X_df = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    y = np.array([1, 2])
+    y = y.astype(dtype=dtype)
+    y_df = _convert_to_dataframe(y, sycl_queue=queue, target_df=dataframe)
+
+    inclin = IncrementalLinearRegression(fit_intercept=fit_intercept)
+    if macro_block is not None:
+        hparams = inclin.get_hyperparameters("fit")
+        hparams.cpu_macro_block = macro_block
+        hparams.gpu_macro_block = macro_block
+    inclin.fit(X_df, y_df)
+
+    y_pred = inclin.predict(X_df)
+
+    tol = 2e-6 if y_pred.dtype == np.float32 else 1e-7
+    assert_allclose(inclin.coef_, [1], atol=tol)
+    if fit_intercept:
+        assert_allclose(inclin.intercept_, [0], atol=tol)
+    assert_allclose(_as_numpy(y_pred), y, atol=tol)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("fit_intercept", [True, False])
+@pytest.mark.parametrize("macro_block", [None, 1024])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_sklearnex_partial_fit_on_gold_data(
+    dataframe, queue, fit_intercept, macro_block, dtype
+):
+    X = np.array([[1], [2], [3], [4]])
+    X = X.astype(dtype=dtype)
+    y = X + 3
+    y = y.astype(dtype=dtype)
+    X_split = np.array_split(X, 2)
+    y_split = np.array_split(y, 2)
+
+    inclin = IncrementalLinearRegression()
+    if macro_block is not None:
+        hparams = inclin.get_hyperparameters("fit")
+        hparams.cpu_macro_block = macro_block
+        hparams.gpu_macro_block = macro_block
+    for i in range(2):
+        X_split_df = _convert_to_dataframe(
+            X_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        y_split_df = _convert_to_dataframe(
+            y_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        inclin.partial_fit(X_split_df, y_split_df)
+
+    X_df = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    y_pred = inclin.predict(X_df)
+
+    assert inclin.n_features_in_ == 1
+    tol = 2e-6 if y_pred.dtype == np.float32 else 1e-7
+    assert_allclose(inclin.coef_, [[1]], atol=tol)
+    if fit_intercept:
+        assert_allclose(inclin.intercept_, 3, atol=tol)
+
+    assert_allclose(_as_numpy(y_pred), y, atol=tol)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("fit_intercept", [True, False])
+@pytest.mark.parametrize("macro_block", [None, 1024])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_sklearnex_partial_fit_multitarget_on_gold_data(
+    dataframe, queue, fit_intercept, macro_block, dtype
+):
+    X = np.array([[1, 1], [1, 2], [2, 2], [2, 3]])
+    X = X.astype(dtype=dtype)
+    y = np.dot(X, [1, 2]) + 3
+    y = y.astype(dtype=dtype)
+    X_split = np.array_split(X, 2)
+    y_split = np.array_split(y, 2)
+
+    inclin = IncrementalLinearRegression()
+    if macro_block is not None:
+        hparams = inclin.get_hyperparameters("fit")
+        hparams.cpu_macro_block = macro_block
+        hparams.gpu_macro_block = macro_block
+    for i in range(2):
+        X_split_df = _convert_to_dataframe(
+            X_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        y_split_df = _convert_to_dataframe(
+            y_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        inclin.partial_fit(X_split_df, y_split_df)
+
+    X_df = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    y_pred = inclin.predict(X_df)
+
+    assert inclin.n_features_in_ == 2
+    tol = 7e-6 if y_pred.dtype == np.float32 else 1e-7
+    assert_allclose(inclin.coef_, [1.0, 2.0], atol=tol)
+    if fit_intercept:
+        assert_allclose(inclin.intercept_, 3.0, atol=tol)
+
+    assert_allclose(_as_numpy(y_pred), y, atol=tol)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("fit_intercept", [True, False])
+@pytest.mark.parametrize("num_samples", [100, 1000])
+@pytest.mark.parametrize("num_features", [5, 10])
+@pytest.mark.parametrize("num_targets", [1, 2])
+@pytest.mark.parametrize("num_blocks", [1, 10])
+@pytest.mark.parametrize("macro_block", [None, 1024])
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+def test_sklearnex_partial_fit_on_random_data(
+    dataframe,
+    queue,
+    fit_intercept,
+    num_samples,
+    num_features,
+    num_targets,
+    num_blocks,
+    macro_block,
+    dtype,
+):
+    seed = 42
+    gen = np.random.default_rng(seed)
+    intercept = gen.random(size=num_targets, dtype=dtype)
+    coef = gen.random(size=(num_targets, num_features), dtype=dtype).T
+
+    X = gen.random(size=(num_samples, num_features), dtype=dtype)
+    if fit_intercept:
+        y = X @ coef + intercept[np.newaxis, :]
+    else:
+        y = X @ coef
+
+    X_split = np.array_split(X, num_blocks)
+    y_split = np.array_split(y, num_blocks)
+
+    inclin = IncrementalLinearRegression(fit_intercept=fit_intercept)
+    if macro_block is not None:
+        hparams = inclin.get_hyperparameters("fit")
+        hparams.cpu_macro_block = macro_block
+        hparams.gpu_macro_block = macro_block
+    for i in range(num_blocks):
+        X_split_df = _convert_to_dataframe(
+            X_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        y_split_df = _convert_to_dataframe(
+            y_split[i], sycl_queue=queue, target_df=dataframe
+        )
+        inclin.partial_fit(X_split_df, y_split_df)
+
+    tol = 1e-4 if inclin.coef_.dtype == np.float32 else 1e-7
+    assert_allclose(coef, inclin.coef_.T, atol=tol)
+
+    if fit_intercept:
+        assert_allclose(intercept, inclin.intercept_, atol=tol)
+
+    X_test = gen.random(size=(num_samples, num_features), dtype=dtype)
+    if fit_intercept:
+        expected_y_pred = X_test @ coef + intercept[np.newaxis, :]
+    else:
+        expected_y_pred = X_test @ coef
+
+    X_test_df = _convert_to_dataframe(X_test, sycl_queue=queue, target_df=dataframe)
+
+    y_pred = inclin.predict(X_test_df)
+
+    assert_allclose(expected_y_pred, _as_numpy(y_pred), atol=tol)

{scikit_learn_intelex-2024.4.0.data → scikit_learn_intelex-2024.6.0.data}/data/Lib/site-packages/sklearnex/linear_model/tests/test_linear.py

@@ -52,7 +52,7 @@ def test_sklearnex_import_linear(dataframe, queue, dtype, macro_block):
     assert "sklearnex" in linreg.__module__
     assert linreg.n_features_in_ == 2
 
-    tol = 1e-5 if X.dtype == np.float32 else 1e-7
+    tol = 1e-5 if _as_numpy(linreg.coef_).dtype == np.float32 else 1e-7
     assert_allclose(_as_numpy(linreg.intercept_), 3.0, rtol=tol)
     assert_allclose(_as_numpy(linreg.coef_), [1.0, 2.0], rtol=tol)
 
@@ -113,5 +113,5 @@ def test_sklearnex_reconstruct_model(dataframe, queue, dtype):
 
     y_pred = linreg.predict(X)
 
-    tol = 1e-5 if X.dtype == np.float32 else 1e-7
+    tol = 1e-5 if _as_numpy(y_pred).dtype == np.float32 else 1e-7
     assert_allclose(gtr, _as_numpy(y_pred), rtol=tol)

{scikit_learn_intelex-2024.4.0.data → scikit_learn_intelex-2024.6.0.data}/data/Lib/site-packages/sklearnex/neighbors/knn_regression.py

@@ -14,6 +14,7 @@
 # limitations under the License.
 # ==============================================================================
 
+from sklearn.metrics import r2_score
 from sklearn.neighbors._regression import (
     KNeighborsRegressor as sklearn_KNeighborsRegressor,
 )
@@ -117,6 +118,23 @@ class KNeighborsRegressor(sklearn_KNeighborsRegressor, KNeighborsDispatchingBase
             X,
         )
 
+    @wrap_output_data
+    def score(self, X, y, sample_weight=None):
+        check_is_fitted(self)
+        if sklearn_check_version("1.0"):
+            self._check_feature_names(X, reset=False)
+        return dispatch(
+            self,
+            "score",
+            {
+                "onedal": self.__class__._onedal_score,
+                "sklearn": sklearn_KNeighborsRegressor.score,
+            },
+            X,
+            y,
+            sample_weight=sample_weight,
+        )
+
     @wrap_output_data
     def kneighbors(self, X=None, n_neighbors=None, return_distance=True):
         check_is_fitted(self)
@@ -184,6 +202,11 @@ class KNeighborsRegressor(sklearn_KNeighborsRegressor, KNeighborsDispatchingBase
             X, n_neighbors, return_distance, queue=queue
         )
 
+    def _onedal_score(self, X, y, sample_weight=None, queue=None):
+        return r2_score(
+            y, self._onedal_predict(X, queue=queue), sample_weight=sample_weight
+        )
+
     def _save_attributes(self):
         self.n_features_in_ = self._onedal_estimator.n_features_in_
         self.n_samples_fit_ = self._onedal_estimator.n_samples_fit_
@@ -196,3 +219,4 @@ class KNeighborsRegressor(sklearn_KNeighborsRegressor, KNeighborsDispatchingBase
     predict.__doc__ = sklearn_KNeighborsRegressor.predict.__doc__
     kneighbors.__doc__ = sklearn_KNeighborsRegressor.kneighbors.__doc__
     radius_neighbors.__doc__ = sklearn_NearestNeighbors.radius_neighbors.__doc__
+    score.__doc__ = sklearn_KNeighborsRegressor.score.__doc__

{scikit_learn_intelex-2024.4.0.data → scikit_learn_intelex-2024.6.0.data}/data/Lib/site-packages/sklearnex/neighbors/tests/test_neighbors.py

@@ -47,9 +47,9 @@ def test_sklearnex_import_knn_regression(dataframe, queue):
     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))
+    pred = _as_numpy(neigh.predict(y_test)).squeeze()
     assert "sklearnex" in neigh.__module__
-    assert_allclose(pred, [0.5])
+    assert_allclose(pred, 0.5)
 
 
 @pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())

{scikit_learn_intelex-2024.4.0.data → scikit_learn_intelex-2024.6.0.data}/data/Lib/site-packages/sklearnex/preview/__init__.py

@@ -14,4 +14,4 @@
 # limitations under the License.
 # ==============================================================================
 
-__all__ = ["cluster", "covariance"]
+__all__ = ["cluster", "covariance", "decomposition"]

scikit_learn_intelex-2024.6.0.data/data/Lib/site-packages/sklearnex/preview/decomposition/__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_pca import IncrementalPCA
+
+__all__ = ["IncrementalPCA"]

scikit_learn_intelex-2024.6.0.data/data/Lib/site-packages/sklearnex/preview/decomposition/incremental_pca.py

@@ -0,0 +1,228 @@
+# ===============================================================================
+# 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.decomposition import IncrementalPCA as sklearn_IncrementalPCA
+from sklearn.utils import check_array, gen_batches
+
+from daal4py.sklearn._n_jobs_support import control_n_jobs
+from daal4py.sklearn._utils import sklearn_check_version
+from onedal.decomposition import IncrementalPCA as onedal_IncrementalPCA
+
+from ..._device_offload import dispatch, wrap_output_data
+from ..._utils import PatchingConditionsChain
+
+
+@control_n_jobs(
+    decorated_methods=["fit", "partial_fit", "transform", "_onedal_finalize_fit"]
+)
+class IncrementalPCA(sklearn_IncrementalPCA):
+
+    def __init__(self, n_components=None, *, whiten=False, copy=True, batch_size=None):
+        super().__init__(
+            n_components=n_components, whiten=whiten, copy=copy, batch_size=batch_size
+        )
+        self._need_to_finalize = False
+        self._need_to_finalize_attrs = {
+            "mean_",
+            "explained_variance_",
+            "explained_variance_ratio_",
+            "n_components_",
+            "components_",
+            "noise_variance_",
+            "singular_values_",
+            "var_",
+        }
+
+    _onedal_incremental_pca = staticmethod(onedal_IncrementalPCA)
+
+    def _onedal_transform(self, X, queue=None):
+        assert hasattr(self, "_onedal_estimator")
+        if self._need_to_finalize:
+            self._onedal_finalize_fit()
+        X = check_array(X, dtype=[np.float64, np.float32])
+        return self._onedal_estimator.predict(X, queue)
+
+    def _onedal_fit_transform(self, X, queue=None):
+        self._onedal_fit(X, queue)
+        return self._onedal_transform(X, queue)
+
+    def _onedal_partial_fit(self, X, check_input=True, queue=None):
+        first_pass = not hasattr(self, "components_")
+
+        if check_input:
+            if sklearn_check_version("1.0"):
+                X = self._validate_data(
+                    X, dtype=[np.float64, np.float32], reset=first_pass
+                )
+            else:
+                X = check_array(
+                    X,
+                    dtype=[np.float64, np.float32],
+                    copy=self.copy,
+                )
+
+        n_samples, n_features = X.shape
+
+        if self.n_components is None:
+            if not hasattr(self, "components_"):
+                self.n_components_ = min(n_samples, n_features)
+            else:
+                self.n_components_ = self.components_.shape[0]
+        elif not self.n_components <= n_features:
+            raise ValueError(
+                "n_components=%r invalid for n_features=%d, need "
+                "more rows than columns for IncrementalPCA "
+                "processing" % (self.n_components, n_features)
+            )
+        elif not self.n_components <= n_samples:
+            raise ValueError(
+                "n_components=%r must be less or equal to "
+                "the batch number of samples "
+                "%d." % (self.n_components, n_samples)
+            )
+        else:
+            self.n_components_ = self.n_components
+
+        if not hasattr(self, "n_samples_seen_"):
+            self.n_samples_seen_ = n_samples
+        else:
+            self.n_samples_seen_ += n_samples
+
+        onedal_params = {"n_components": self.n_components_, "whiten": self.whiten}
+
+        if not hasattr(self, "_onedal_estimator"):
+            self._onedal_estimator = self._onedal_incremental_pca(**onedal_params)
+        self._onedal_estimator.partial_fit(X, queue)
+        self._need_to_finalize = True
+
+    def _onedal_finalize_fit(self):
+        assert hasattr(self, "_onedal_estimator")
+        self._onedal_estimator.finalize_fit()
+        self._need_to_finalize = False
+
+    def _onedal_fit(self, X, queue=None):
+        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], copy=self.copy)
+        else:
+            X = check_array(
+                X,
+                dtype=[np.float64, np.float32],
+                copy=self.copy,
+            )
+
+        n_samples, n_features = X.shape
+
+        if self.batch_size is None:
+            self.batch_size_ = 5 * n_features
+        else:
+            self.batch_size_ = self.batch_size
+
+        self.n_samples_seen_ = 0
+        if hasattr(self, "_onedal_estimator"):
+            self._onedal_estimator._reset()
+
+        for batch in gen_batches(n_samples, self.batch_size_):
+            X_batch = X[batch]
+            self._onedal_partial_fit(X_batch, queue=queue)
+
+        self._onedal_finalize_fit()
+
+        return self
+
+    def _onedal_supported(self, method_name, *data):
+        patching_status = PatchingConditionsChain(
+            f"sklearn.decomposition.{self.__class__.__name__}.{method_name}"
+        )
+        return patching_status
+
+    _onedal_cpu_supported = _onedal_supported
+    _onedal_gpu_supported = _onedal_supported
+
+    def __getattr__(self, attr):
+        if attr in self._need_to_finalize_attrs:
+            if hasattr(self, "_onedal_estimator"):
+                if self._need_to_finalize:
+                    self._onedal_finalize_fit()
+                return getattr(self._onedal_estimator, attr)
+            else:
+                raise AttributeError(
+                    f"'{self.__class__.__name__}' object has no attribute '{attr}'"
+                )
+        if attr in self.__dict__:
+            return self.__dict__[attr]
+
+        raise AttributeError(
+            f"'{self.__class__.__name__}' object has no attribute '{attr}'"
+        )
+
+    def partial_fit(self, X, y=None, check_input=True):
+        dispatch(
+            self,
+            "partial_fit",
+            {
+                "onedal": self.__class__._onedal_partial_fit,
+                "sklearn": sklearn_IncrementalPCA.partial_fit,
+            },
+            X,
+            check_input=check_input,
+        )
+        return self
+
+    def fit(self, X, y=None):
+        dispatch(
+            self,
+            "fit",
+            {
+                "onedal": self.__class__._onedal_fit,
+                "sklearn": sklearn_IncrementalPCA.fit,
+            },
+            X,
+        )
+        return self
+
+    @wrap_output_data
+    def transform(self, X):
+        return dispatch(
+            self,
+            "transform",
+            {
+                "onedal": self.__class__._onedal_transform,
+                "sklearn": sklearn_IncrementalPCA.transform,
+            },
+            X,
+        )
+
+    @wrap_output_data
+    def fit_transform(self, X, y=None, **fit_params):
+        return dispatch(
+            self,
+            "fit_transform",
+            {
+                "onedal": self.__class__._onedal_fit_transform,
+                "sklearn": sklearn_IncrementalPCA.fit_transform,
+            },
+            X,
+        )
+
+    __doc__ = sklearn_IncrementalPCA.__doc__
+    fit.__doc__ = sklearn_IncrementalPCA.fit.__doc__
+    fit_transform.__doc__ = sklearn_IncrementalPCA.fit_transform.__doc__
+    transform.__doc__ = sklearn_IncrementalPCA.transform.__doc__
+    partial_fit.__doc__ = sklearn_IncrementalPCA.partial_fit.__doc__