Relax UMAP version constraint

conda/environments/all_cuda-129_arch-aarch64.yaml

@@ -81,5 +81,5 @@ dependencies:
 - statsmodels
 - sysroot_linux-aarch64==2.28
 - treelite>=4.6.1,<5.0.0
-- umap-learn==0.5.7
+- umap-learn>=0.5.7,<0.5.12
 name: all_cuda-129_arch-aarch64

conda/environments/all_cuda-129_arch-x86_64.yaml

@@ -81,5 +81,5 @@ dependencies:
 - statsmodels
 - sysroot_linux-64==2.28
 - treelite>=4.6.1,<5.0.0
-- umap-learn==0.5.7
+- umap-learn>=0.5.7,<0.5.12
 name: all_cuda-129_arch-x86_64

conda/environments/all_cuda-131_arch-aarch64.yaml

@@ -81,5 +81,5 @@ dependencies:
 - statsmodels
 - sysroot_linux-aarch64==2.28
 - treelite>=4.6.1,<5.0.0
-- umap-learn==0.5.7
+- umap-learn>=0.5.7,<0.5.12
 name: all_cuda-131_arch-aarch64

conda/environments/all_cuda-131_arch-x86_64.yaml

@@ -81,5 +81,5 @@ dependencies:
 - statsmodels
 - sysroot_linux-64==2.28
 - treelite>=4.6.1,<5.0.0
-- umap-learn==0.5.7
+- umap-learn>=0.5.7,<0.5.12
 name: all_cuda-131_arch-x86_64

dependencies.yaml

@@ -455,9 +455,11 @@ dependencies:
           - matrix: {dependencies: "oldest"}
             packages:
               - scikit-learn==1.5.0
+              - umap-learn==0.5.7
           - matrix: {dependencies: "intermediate"}
             packages:
               - scikit-learn==1.7.2
+              - umap-learn==0.5.8
               - ipython==8.10.0
           - matrix: {dependencies: "nightly"}
             packages:
@@ -512,7 +514,7 @@ dependencies:
           - pytest-xdist
           - seaborn
           - statsmodels
-          - umap-learn==0.5.7
+          - umap-learn>=0.5.7,<0.5.12
           - pynndescent
       - output_types: conda
         packages:

docs/source/supported_versions.rst

@@ -36,7 +36,7 @@ The following dependencies are optional and provide additional functionality:
 
 * **xgboost**: >=2.1.0 (for gradient boosting algorithms)
 * **hdbscan**: >=0.8.39,<0.8.40 (for hierarchical density-based clustering)
-* **umap-learn**: ==0.5.7 (for dimensionality reduction)
+* **umap-learn**: >=0.5.7,<0.5.12 (for dimensionality reduction)
 * **pynndescent**: (for approximate nearest neighbor search)
 
 RAPIDS Dependencies

python/cuml/cuml/accel/_overrides/umap.py

@@ -3,30 +3,57 @@
 # SPDX-License-Identifier: Apache-2.0
 #
 
+from packaging.version import Version
+
 import cuml.manifold
 from cuml.accel.estimator_proxy import ProxyBase
 
+try:
+    import umap as _umap_module
+
+    _UMAP_LT_058 = Version(_umap_module.__version__) < Version("0.5.8")
+except ImportError:
+    _UMAP_LT_058 = False
+
 __all__ = ("UMAP",)
 
 
 class UMAP(ProxyBase):
     _gpu_class = cuml.manifold.UMAP
 
-    def _gpu_fit(self, X, y=None, force_all_finite=True, **kwargs):
-        # **kwargs is here for signature compatibility - umap.UMAP has them,
-        # but ignores all but the ones named here.
-        # TODO: cuml.UMAP currently doesn't handle non-finite inputs.
-        # force_alL_finite is in here for _signature_ compatibility
-        # with umap.UMAP, but we don't properly implement it (yet).
-        return self._gpu.fit(X, y=y)
-
-    def _gpu_fit_transform(self, X, y=None, force_all_finite=True, **kwargs):
-        # **kwargs is here for signature compatibility - umap.UMAP has them,
-        # but ignores all but the ones named here.
-        return self._gpu.fit_transform(X, y=y)
-
-    def _gpu_transform(self, X, force_all_finite=True):
-        return self._gpu.transform(X)
+    if _UMAP_LT_058:
+        # We support the old signature for backwards compatibility prior to umap-learn 0.5.8
+
+        def _gpu_fit(self, X, y=None, force_all_finite=True, **kwargs):
+            return self._gpu.fit(X, y=y)
+
+        def _gpu_fit_transform(
+            self, X, y=None, force_all_finite=True, **kwargs
+        ):
+            return self._gpu.fit_transform(X, y=y)
+
+        def _gpu_transform(self, X, force_all_finite=True):
+            return self._gpu.transform(X)
+
+    else:
+
+        def _gpu_fit(self, X, y=None, ensure_all_finite=True, **kwargs):
+            # **kwargs is here for signature compatibility - umap.UMAP has them,
+            # but ignores all but the ones named here.
+            # TODO: cuml.UMAP currently doesn't handle non-finite inputs.
+            # ensure_all_finite is in here for _signature_ compatibility
+            # with umap.UMAP, but we don't properly implement it (yet).
+            return self._gpu.fit(X, y=y)
+
+        def _gpu_fit_transform(
+            self, X, y=None, ensure_all_finite=True, **kwargs
+        ):
+            # **kwargs is here for signature compatibility - umap.UMAP has them,
+            # but ignores all but the ones named here.
+            return self._gpu.fit_transform(X, y=y)
+
+        def _gpu_transform(self, X, ensure_all_finite=True):
+            return self._gpu.transform(X)
 
     def _gpu_inverse_transform(self, X):
         return self._gpu.inverse_transform(X)

python/cuml/cuml/accel/pytest_plugin.py

@@ -43,28 +43,46 @@ def pytest_addoption(parser):
     )
 
 
+def _evaluate_single_condition(condition_str: str) -> bool:
+    """Evaluate a single package version condition.
+
+    Args:
+        condition_str: String in format 'package[comparison]version',
+                      e.g. 'scikit-learn>=1.5.2'
+
+    Returns:
+        bool: True if the condition is met, False otherwise
+    """
+    try:
+        req = Requirement(condition_str.strip())
+        installed_version = version(req.name)
+        return req.specifier.contains(installed_version, prereleases=True)
+    except Exception:
+        return False
+
+
 def create_version_condition(condition_str: str) -> bool:
     """Evaluate a version condition immediately.
 
+    Supports a single package specifier or multiple specifiers joined by 'and',
+    in which case all clauses must be satisfied.
+
     Args:
-        condition_str: String in format 'package[comparison]version'
-                      For example:
+        condition_str: A version condition string. Examples:
                       - 'scikit-learn>=1.5.2'
                       - 'numpy<2.0.0'
-                      - 'pandas==2.1.0'
+                      - 'umap-learn<=0.5.8 and scikit-learn>=1.6'
 
     Returns:
-        bool: True if the condition is met, False otherwise
+        bool: True if all conditions are met, False otherwise
     """
     if not condition_str:
         return True
 
-    try:
-        req = Requirement(condition_str)
-        installed_version = version(req.name)
-        return req.specifier.contains(installed_version, prereleases=True)
-    except Exception:
-        return False
+    return all(
+        _evaluate_single_condition(clause)
+        for clause in condition_str.split(" and ")
+    )
 
 
 def pytest_collection_modifyitems(config, items):

python/cuml/cuml_accel_tests/integration/test_umap.py

@@ -4,25 +4,10 @@
 #
 
 import pytest
-import sklearn
-from packaging.version import Version
 from sklearn.datasets import make_swiss_roll
 from sklearn.manifold import trustworthiness
 from umap import UMAP
 
-if Version(sklearn.__version__) >= Version("1.8.0.dev0"):
-    pytest.skip("umap requires sklearn < 1.8.0.dev0", allow_module_level=True)
-
-# Ignore FutureWarning from third-party umap-learn package calling
-# sklearn.utils.validation.check_array with deprecated 'force_all_finite'
-# parameter. This is not in cuml's control. Note: this will break when
-# sklearn 1.8 removes the deprecated parameter entirely - umap-learn will
-# need to be updated at that point.
-# See also https://github.com/lmcinnes/umap/issues/1174
-pytestmark = pytest.mark.filterwarnings(
-    "ignore:'force_all_finite' was renamed to 'ensure_all_finite':FutureWarning:sklearn"
-)
-
 
 @pytest.fixture(scope="module")
 def manifold_data():
@@ -65,6 +50,7 @@ def test_umap_min_dist(manifold_data, min_dist):
         "russellrao",
         "kulsinski",
         "dice",
+        # Require 2D data for these metrics
         "wminkowski",
         "mahalanobis",
         "haversine",
@@ -82,7 +68,7 @@ def test_umap_min_dist(manifold_data, min_dist):
 def test_umap_metric(manifold_data, metric):
     X = manifold_data
     # haversine only works for 2D data
-    if metric == "haversine":
+    if metric in ["haversine", "wminkowski", "mahalanobis"]:
         X = X[:, :2]
 
     umap = UMAP(metric=metric, random_state=42)

python/cuml/cuml_accel_tests/test_core.py

@@ -1,10 +1,12 @@
-# SPDX-FileCopyrightText: Copyright (c) 2025, NVIDIA CORPORATION.
+# SPDX-FileCopyrightText: Copyright (c) 2025-2026, NVIDIA CORPORATION.
 # SPDX-License-Identifier: Apache-2.0
 import importlib
+import importlib.metadata
 import multiprocessing
 from inspect import Parameter, signature
 
 import pytest
+from packaging.version import Version
 
 import cuml.accel
 from cuml.accel.estimator_proxy import ProxyBase
@@ -80,17 +82,19 @@ def iter_proxy_class_methods():
             if not name.startswith("_") and callable(
                 getattr(cls._cpu_class, name)
             ):
-                # XXX: xfail umap.UMAP.get_feature_names_out for now
+                # XXX: xfail umap.UMAP.get_feature_names_out for umap-learn < 0.5.8
                 if (
                     cls._cpu_class.__name__ == "UMAP"
                     and name == "get_feature_names_out"
+                    and Version(importlib.metadata.version("umap-learn"))
+                    < Version("0.5.8")
                 ):
                     yield pytest.param(
                         cls,
                         name,
                         marks=[
                             pytest.mark.xfail(
-                                reason="umap-learn <= 0.5.7 doesn't implement `get_feature_names_out` properly",
+                                reason="umap-learn < 0.5.8 doesn't implement `get_feature_names_out` properly",
                                 strict=True,
                             )
                         ],

python/cuml/cuml_accel_tests/upstream/README.md

@@ -179,6 +179,8 @@ When tests fail only on specific versions of a dependency, use the `--condition`
 
 This ensures xfails only apply to the relevant versions, keeping the test suite accurate across different dependency combinations.
 
+You can combine multiple version requirements in a condition using "and". For example: `umap-learn<=0.5.8 and scikit-learn>=1.6` will only xfail tests when both package constraints are met.
+
 ### Handling Unmatched Test IDs
 
 The pytest plugin validates that all test IDs in the xfail list correspond to actual tests. When tests don't exist, a `UnmatchedXfailTests` warning is issued.

python/cuml/cuml_accel_tests/upstream/umap/run-tests.sh

@@ -12,30 +12,18 @@
 
 set -eu
 
-UMAP_TAG="release-0.5.7"
-
-# Skip tests for scikit-learn >= 1.8 -- umap-learn is not compatible with scikit-learn 1.8 yet
-python -c "
-import sys
-from packaging.version import Version
-import sklearn
-sys.exit(
-    int(
-        Version(sklearn.__version__) >= Version('1.8')
-    )
-)
-" || {
-    echo "Skipping umap tests for scikit-learn >= 1.8"
-    exit 0
-}
+UMAP_VERSION=$(python -c "import umap; print(umap.__version__)")
+UMAP_TAG="release-${UMAP_VERSION}"
 
 THIS_DIRECTORY=$( cd "$(dirname "${BASH_SOURCE[0]}")" ; pwd -P )
 UMAP_REPO="${THIS_DIRECTORY}/umap-upstream"
 
-# Shallow clone the tag if not already cloned
+# Clone if not already present, then check out the matching tag
 if [ ! -d "$UMAP_REPO" ]; then
-    git clone --branch $UMAP_TAG --depth 1 "https://github.com/lmcinnes/umap.git" "$UMAP_REPO"
+    git clone "https://github.com/lmcinnes/umap.git" "$UMAP_REPO"
 fi
+git -C "$UMAP_REPO" fetch --tags
+git -C "$UMAP_REPO" checkout "$UMAP_TAG"
 
 # Run upstream tests
 pytest -p cuml.accel \

python/cuml/cuml_accel_tests/upstream/umap/xfail-list.yaml

@@ -42,6 +42,6 @@
   - "umap.tests.test_umap_ops::test_umap_transform_embedding_stability"
 - reason: Test fails with newer sklearn
   marker: cuml_accel_sklearn_pin
-  condition: scikit-learn>=1.6
+  condition: umap-learn<=0.5.8 and scikit-learn>=1.6
   tests:
   - "umap.tests.test_umap_validation_params::test_umap_custom_distance_w_grad"

python/cuml/cuml_accel_tests/upstream/xfail_manager.py

@@ -31,7 +31,7 @@
 from typing import Any, Dict, List, Optional, Union
 
 import yaml
-from packaging.requirements import Requirement
+from packaging.requirements import InvalidRequirement, Requirement
 
 
 class QuoteTestID(str):
@@ -358,13 +358,14 @@ def validate_conditions(self) -> List[str]:
         errors = []
         for i, group in enumerate(self.groups):
             if group.condition:
-                try:
-                    Requirement(group.condition)
-                except Exception as e:
-                    errors.append(
-                        f"Group {i} has invalid condition "
-                        f"'{group.condition}': {e}"
-                    )
+                for clause in group.condition.split(" and "):
+                    try:
+                        Requirement(clause.strip())
+                    except InvalidRequirement as e:
+                        errors.append(
+                            f"Group {i} has invalid condition "
+                            f"'{group.condition}': {e}"
+                        )
         return errors
 
     def _merge_identical_groups(self) -> int:

python/cuml/pyproject.toml

@@ -126,7 +126,7 @@ test = [
     "seaborn",
     "skl2onnx",
     "statsmodels",
-    "umap-learn==0.5.7",
+    "umap-learn>=0.5.7,<0.5.12",
     "xgboost>=2.1.0",
 ] # This list was generated by `rapids-dependency-file-generator`. To make changes, edit ../../dependencies.yaml and run `rapids-dependency-file-generator`.
 dask = [

python/cuml/tests/test_sklearn_import_export.py

@@ -477,17 +477,6 @@ def test_svc_multiclass_unsupported(random_state):
 
 @pytest.mark.parametrize("sparse", [False, True])
 @pytest.mark.parametrize("supervised", [False, True])
-@pytest.mark.skipif(SKLEARN_18, reason="umap requires sklearn < 1.8.0")
-# Ignore FutureWarning from third-party umap-learn package calling
-# sklearn.utils.validation.check_array with deprecated 'force_all_finite'
-# parameter. This is not in cuml's control. Note: this will break when
-# sklearn 1.8 removes the deprecated parameter entirely - umap-learn will
-# need to be updated at that point.
-# See also https://github.com/lmcinnes/umap/issues/1174
-@pytest.mark.filterwarnings(
-    "ignore:'force_all_finite' was renamed to "
-    "'ensure_all_finite':FutureWarning:sklearn"
-)
 def test_umap(random_state, sparse, supervised):
     n_neighbors = 10
     X, y = make_blobs(n_samples=200, random_state=random_state)