scikit-learn-intelex 2024.3.0__py310-none-win_amd64.whl → 2024.5.0__py310-none-win_amd64.whl

scikit_learn_intelex-2024.5.0.data/data/Lib/site-packages/sklearnex/utils/_namespace.py

@@ -0,0 +1,97 @@
+# ==============================================================================
+# 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

scikit_learn_intelex-2024.5.0.data/data/Lib/site-packages/sklearnex/utils/tests/test_finite.py

@@ -0,0 +1,89 @@
+# ==============================================================================
+# 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 time
+
+import numpy as np
+import numpy.random as rand
+import pytest
+from numpy.testing import assert_raises
+
+from sklearnex.utils import _assert_all_finite
+
+
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+@pytest.mark.parametrize(
+    "shape",
+    [
+        [16, 2048],
+        [
+            2**16 + 3,
+        ],
+        [1000, 1000],
+    ],
+)
+@pytest.mark.parametrize("allow_nan", [False, True])
+def test_sum_infinite_actually_finite(dtype, shape, allow_nan):
+    X = np.array(shape, dtype=dtype)
+    X.fill(np.finfo(dtype).max)
+    _assert_all_finite(X, allow_nan=allow_nan)
+
+
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+@pytest.mark.parametrize(
+    "shape",
+    [
+        [16, 2048],
+        [
+            2**16 + 3,
+        ],
+        [1000, 1000],
+    ],
+)
+@pytest.mark.parametrize("allow_nan", [False, True])
+@pytest.mark.parametrize("check", ["inf", "NaN", None])
+@pytest.mark.parametrize("seed", [0, int(time.time())])
+def test_assert_finite_random_location(dtype, shape, allow_nan, check, seed):
+    rand.seed(seed)
+    X = rand.uniform(high=np.finfo(dtype).max, size=shape).astype(dtype)
+
+    if check:
+        loc = rand.randint(0, X.size - 1)
+        X.reshape((-1,))[loc] = float(check)
+
+    if check is None or (allow_nan and check == "NaN"):
+        _assert_all_finite(X, allow_nan=allow_nan)
+    else:
+        assert_raises(ValueError, _assert_all_finite, X, allow_nan=allow_nan)
+
+
+@pytest.mark.parametrize("dtype", [np.float32, np.float64])
+@pytest.mark.parametrize("allow_nan", [False, True])
+@pytest.mark.parametrize("check", ["inf", "NaN", None])
+@pytest.mark.parametrize("seed", [0, int(time.time())])
+def test_assert_finite_random_shape_and_location(dtype, allow_nan, check, seed):
+    lb, ub = 32768, 1048576  # lb is a patching condition, ub 2^20
+    rand.seed(seed)
+    X = rand.uniform(high=np.finfo(dtype).max, size=rand.randint(lb, ub)).astype(dtype)
+
+    if check:
+        loc = rand.randint(0, X.size - 1)
+        X[loc] = float(check)
+
+    if check is None or (allow_nan and check == "NaN"):
+        _assert_all_finite(X, allow_nan=allow_nan)
+    else:
+        assert_raises(ValueError, _assert_all_finite, X, allow_nan=allow_nan)