Merge branch 'branch-25.04' into r2-score-compat

Author: dantegd

.github/CODEOWNERS

@@ -12,11 +12,11 @@ CMakeLists.txt @rapidsai/cuml-cmake-codeowners
 #CI code owners
 /.github/                @rapidsai/ci-codeowners
 /ci/                     @rapidsai/ci-codeowners
-/.pre-commit-config.yaml @rapidsai/ci-codeowners
 
 #packaging code owners
-/.devcontainer/    @rapidsai/packaging-codeowners
-/conda/            @rapidsai/packaging-codeowners
-/dependencies.yaml @rapidsai/packaging-codeowners
-/build.sh          @rapidsai/packaging-codeowners
-pyproject.toml     @rapidsai/packaging-codeowners
+/.pre-commit-config.yaml @rapidsai/packaging-codeowners
+/.devcontainer/          @rapidsai/packaging-codeowners
+/conda/                  @rapidsai/packaging-codeowners
+dependencies.yaml        @rapidsai/packaging-codeowners
+/build.sh                @rapidsai/packaging-codeowners
+pyproject.toml           @rapidsai/packaging-codeowners

.pre-commit-config.yaml

@@ -65,7 +65,7 @@ repos:
             pass_filenames: false
             language: python
     - repo: https://github.com/rapidsai/pre-commit-hooks
-      rev: v0.4.0
+      rev: v0.5.0
       hooks:
         - id: verify-copyright
           files: |
@@ -80,6 +80,8 @@ repos:
                 cpp/src/tsne/cannylab/bh[.]cu$|
                 python/cuml/cuml/_thirdparty
         - id: verify-alpha-spec
+        - id: verify-codeowners
+          args: [--fix, --project-prefix=cuml]
     - repo: https://github.com/rapidsai/dependency-file-generator
       rev: v1.17.0
       hooks:

ci/run_cuml_singlegpu_accel_pytests.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
-# Copyright (c) 2024, NVIDIA CORPORATION.
+# Copyright (c) 2024-2025, NVIDIA CORPORATION.
 
 # Support invoking run_cuml_singlegpu_pytests.sh outside the script directory
 cd "$(dirname "$(realpath "${BASH_SOURCE[0]}")")"/../python/cuml/cuml/tests/experimental/accel
 
-python -m pytest -p cuml.experimental.accel --cache-clear "$@" .
+python -m pytest -p cuml.accel --cache-clear "$@" .

python/cuml/cuml/accel/__init__.py

@@ -0,0 +1,21 @@
+#
+# Copyright (c) 2025, NVIDIA 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 cuml.experimental.accel import (
+    install,
+    load_ipython_extension,
+    pytest_load_initial_conftests,
+)

python/cuml/cuml/accel/__main__.py

@@ -0,0 +1,21 @@
+#
+# Copyright (c) 2025, NVIDIA 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 cuml.experimental.accel.__main__ import main
+
+
+if __name__ == "__main__":
+    main()

python/cuml/cuml/internals/api_decorators.py

@@ -55,27 +55,32 @@ def _find_arg(sig, arg_name, default_position):
 
     # Check for default name in input args
     if arg_name in sig.parameters:
-        return arg_name, params.index(arg_name)
+        param = sig.parameters[arg_name]
+        return arg_name, params.index(arg_name), param.default
     # Otherwise use argument in list by position
     elif arg_name is ...:
         index = int(_has_self(sig)) + default_position
-        return params[index], index
+        param = params[index]
+        return param, index, sig.parameters[param].default
     else:
         raise ValueError(f"Unable to find parameter '{arg_name}'.")
 
 
-def _get_value(args, kwargs, name, index):
+def _get_value(args, kwargs, name, index, default_value):
     """Determine value for a given set of args, kwargs, name and index."""
     try:
         return kwargs[name]
     except KeyError:
         try:
             return args[index]
         except IndexError:
-            raise IndexError(
-                f"Specified arg idx: {index}, and argument name: {name}, "
-                "were not found in args or kwargs."
-            )
+            if default_value is not inspect._empty:
+                return default_value
+            else:
+                raise IndexError(
+                    f"Specified arg idx: {index}, and argument name: {name}, "
+                    "were not found in args or kwargs."
+                )
 
 
 def _make_decorator_function(
@@ -166,7 +171,7 @@ def wrapper(*args, **kwargs):
                         if self_val is None:
                             assert input_val is not None
                             out_type = iu.determine_array_type(input_val)
-                        elif input_val is None:
+                        elif input_val is None or input_val is inspect._empty:
                             out_type = self_val.output_type
                             if out_type == "input":
                                 out_type = self_val._input_type

python/cuml/cuml/internals/input_utils.py

@@ -1,5 +1,5 @@
 #
-# Copyright (c) 2019-2024, NVIDIA CORPORATION.
+# Copyright (c) 2019-2025, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -530,6 +530,8 @@ def input_to_host_array(
 
 
 def input_to_host_array_with_sparse_support(X):
+    if X is None:
+        return None
     try:
         if scipy_sparse.isspmatrix(X):
             return X

python/cuml/cuml/neighbors/nearest_neighbors.pyx

@@ -1,5 +1,5 @@
 #
-# Copyright (c) 2019-2024, NVIDIA CORPORATION.
+# Copyright (c) 2019-2025, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -28,6 +28,7 @@ import math
 import cuml.internals
 from cuml.internals.base import UniversalBase
 from cuml.common.array_descriptor import CumlArrayDescriptor
+from cuml.internals import api_base_return_generic
 from cuml.internals.array import CumlArray
 from cuml.internals.array_sparse import SparseCumlArray
 from cuml.common.doc_utils import generate_docstring
@@ -691,6 +692,9 @@ class NearestNeighbors(UniversalBase,
             if out_type in {'cupy', 'numpy', 'numba'}:
                 I_ndarr = I_ndarr[:, 1:]
                 D_ndarr = D_ndarr[:, 1:]
+            elif out_type == "cuml":
+                I_ndarr = CumlArray.from_input(I_ndarr[:, 1:], force_contiguous=True)
+                D_ndarr = CumlArray.from_input(D_ndarr[:, 1:], force_contiguous=True)
             else:
                 I_ndarr.drop(I_ndarr.columns[0], axis=1)
                 D_ndarr.drop(D_ndarr.columns[0], axis=1)

python/cuml/cuml/tests/experimental/accel/test_basic_estimators.py

@@ -173,6 +173,18 @@ def test_proxy_facade():
             assert original_value == proxy_value
 
 
+def test_defaults_args_only_methods():
+    # Check that estimator methods that take no arguments work
+    # These are slightly weird because basically everything else takes
+    # a X as input.
+    X = np.random.rand(1000, 3)
+    y = X[:, 0] + np.sin(6 * np.pi * X[:, 1]) + 0.1 * np.random.randn(1000)
+
+    nn = NearestNeighbors(metric="chebyshev", n_neighbors=3)
+    nn.fit(X[:, 0].reshape((-1, 1)), y)
+    nn.kneighbors()
+
+
 def test_kernel_ridge():
     rng = np.random.RandomState(42)
 

python/cuml/cuml/tests/experimental/accel/test_pipeline.py

@@ -31,13 +31,15 @@
     KNeighborsClassifier,
     KNeighborsRegressor,
 )
+from sklearn.preprocessing import Normalizer
 from sklearn.pipeline import Pipeline, make_pipeline
 from sklearn.datasets import make_classification, make_regression
 from sklearn.model_selection import train_test_split
 from sklearn.metrics import accuracy_score, mean_squared_error
 from umap import UMAP
 import hdbscan
 import numpy as np
+import scipy as sp
 
 
 @pytest.fixture
@@ -154,3 +156,16 @@ def test_automatic_step_naming():
 
     assert "pca" in pipeline.named_steps
     assert "logisticregression" in pipeline.named_steps
+
+
+def test_pipeline_adding_none_value_as_labels(classification_data):
+    X_train, _, _, _ = classification_data
+    X_train = sp.sparse.csr_matrix(X_train)
+
+    # Since cuML's TruncatedSVD does not handle sparse data,
+    # the task will be automatically dispatched to Scikit-Learn.
+    # If no labels are provided, Scikit-Learn's pipeline adds
+    # y=None as the default labels.
+
+    pipeline = make_pipeline(TruncatedSVD(n_components=20))
+    pipeline.fit_transform(X_train)