scikit-learn-intelex 2025.10.0__py313-none-win_amd64.whl

scikit_learn_intelex-2025.10.0.data/data/Lib/site-packages/sklearnex/decomposition/tests/test_pca.py

@@ -0,0 +1,164 @@
+# ===============================================================================
+# 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 numpy as np
+import pytest
+import scipy.sparse as sp
+from numpy.testing import assert_allclose
+from sklearn.base import clone
+from sklearn.datasets import load_iris
+
+from daal4py.sklearn._utils import daal_check_version, sklearn_check_version
+from onedal.tests.utils._dataframes_support import (
+    _as_numpy,
+    _convert_to_dataframe,
+    get_dataframes_and_queues,
+)
+from sklearnex.decomposition import PCA
+
+
+@pytest.fixture
+def hyperparameters(request):
+    hparams = PCA.get_hyperparameters("fit")
+
+    def restore_hyperparameters():
+        if daal_check_version((2025, "P", 700)):
+            PCA.reset_hyperparameters("fit")
+
+    request.addfinalizer(restore_hyperparameters)
+    return hparams
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+@pytest.mark.parametrize("macro_block", [None, 2])
+@pytest.mark.parametrize("grain_size", [None, 2])
+def test_sklearnex_import(hyperparameters, dataframe, queue, macro_block, grain_size):
+
+    X = [[-1, -1], [-2, -1], [-3, -2], [1, 1], [2, 1], [3, 2]]
+    X = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+    X_transformed_expected = [
+        [-1.38340578, -0.2935787],
+        [-2.22189802, 0.25133484],
+        [-3.6053038, -0.04224385],
+        [1.38340578, 0.2935787],
+        [2.22189802, -0.25133484],
+        [3.6053038, 0.04224385],
+    ]
+
+    if daal_check_version((2025, "P", 700)):
+        if macro_block is not None:
+            if queue and queue.sycl_device.is_gpu:
+                pytest.skip("Test for CPU-only functionality")
+            hyperparameters.cpu_macro_block = macro_block
+        if grain_size is not None:
+            if queue and queue.sycl_device.is_gpu:
+                pytest.skip("Test for CPU-only functionality")
+            hyperparameters.cpu_grain_size = grain_size
+
+    pca = PCA(n_components=2, svd_solver="covariance_eigh")
+
+    pca.fit(X)
+    X_transformed = pca.transform(X)
+    X_fit_transformed = PCA(n_components=2, svd_solver="covariance_eigh").fit_transform(X)
+
+    if daal_check_version((2024, "P", 100)):
+        assert "sklearnex" in pca.__module__
+        assert hasattr(pca, "_onedal_estimator")
+    else:
+        assert "daal4py" in pca.__module__
+
+    tol = 1e-5 if _as_numpy(X_transformed).dtype == np.float32 else 1e-7
+    assert_allclose([6.30061232, 0.54980396], _as_numpy(pca.singular_values_))
+    assert_allclose(X_transformed_expected, _as_numpy(X_transformed), rtol=tol)
+    assert_allclose(X_transformed_expected, _as_numpy(X_fit_transformed), rtol=tol)
+
+
+@pytest.mark.skipif(
+    not daal_check_version((2025, "P", 700)),
+    reason="Functionality introduced in a later version",
+)
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues())
+def test_non_batched_covariance(hyperparameters, dataframe, queue):
+    if queue and queue.sycl_device.is_gpu:
+        pytest.skip("Test for CPU-only functionality")
+
+    from sklearnex.decomposition import PCA
+
+    # This generates a random matrix with non-independent columns
+    rng = np.random.default_rng(seed=123)
+    S = rng.standard_normal(size=(6, 5))
+    S = S.T @ S
+    mu = rng.standard_normal(size=5)
+    X = rng.multivariate_normal(mu, S, size=20)
+
+    hyperparameters.cpu_max_cols_batched = np.iinfo(np.int32).max
+    res_batched = PCA().fit(X).components_
+
+    hyperparameters.cpu_max_cols_batched = 1
+    res_non_batched = PCA().fit(X).components_
+
+    assert_allclose(res_non_batched, res_batched)
+
+
+@pytest.mark.parametrize("dataframe,queue", get_dataframes_and_queues("numpy,dpctl"))
+def test_changed_estimated_attributes(with_array_api, dataframe, queue):
+    # check that attributes necessary for the PCA onedal estimator match
+    # changes occurring in the sklearnex estimator
+    X, y = load_iris(return_X_y=True)
+
+    X_0 = _convert_to_dataframe(X[y == 0], sycl_queue=queue, target_df=dataframe)
+    X = _convert_to_dataframe(X, sycl_queue=queue, target_df=dataframe)
+
+    est0 = PCA(n_components=4).fit(X_0)
+    est = clone(est0)
+    assert not hasattr(est, "_onedal_estimator")
+    est.fit(X)
+    assert not np.array_equal(_as_numpy(est.transform(X)), _as_numpy(est0.transform(X)))
+
+    # copy over parameters necessary for transform
+    est.mean_ = est0.mean_
+    est.components_ = est0.components_
+    est.explained_variance_ = est0.explained_variance_
+    est.n_components_ = est0.n_components_  # is trivial but exercises the logic
+
+    assert np.array_equal(_as_numpy(est.transform(X)), _as_numpy(est0.transform(X)))
+
+
+@pytest.mark.skipif(
+    not sklearn_check_version("1.5"),
+    reason='svd_solver="auto" does not support sparse inputs',
+)
+@pytest.mark.allow_sklearn_fallback
+def test_create_model_behavior():
+    # verify that fit fallbacks does not break ``transform`` as the oneDAL
+    # model is generated JIT
+
+    X, _ = load_iris(return_X_y=True)
+    # generate a onedal estimator
+    est = PCA(n_components=3)
+    X_trans = est.fit_transform(X)
+
+    # force data to sparse for a fallback to sklearn
+    X_sp = sp.csr_matrix(X)
+    est.fit(X_sp)
+    # In the case of a fallback, the model should be set to none by clobbered
+    # fitted attributes
+    assert est._onedal_estimator._onedal_model is None
+
+    X_trans_sparse = est.transform(X)
+    # use allclose as data was fit with sklearn and onedal on the same data
+    # but using different backends
+    assert_allclose(X_trans, X_trans_sparse)