FIX missing imports and other small fixes

python/cuml/cuml/internals/base.pyx

@@ -37,6 +37,7 @@ except ImportError:
 
 import cuml
 import cuml.common
+from cuml.common.sparse_utils import is_sparse
 import cuml.internals.logger as logger
 import cuml.internals
 import cuml.internals.input_utils
@@ -47,6 +48,7 @@ from cuml.internals.input_utils import (
     determine_array_type,
     input_to_cuml_array,
     input_to_host_array,
+    input_to_host_array_with_sparse_support,
     is_array_like
 )
 from cuml.internals.memory_utils import determine_array_memtype
@@ -676,7 +678,9 @@ class UniversalBase(Base):
 
     def args_to_cpu(self, *args, **kwargs):
         # put all the args on host
-        new_args = tuple(input_to_host_array_with_sparse_support(arg) for arg in args)
+        new_args = tuple(
+            input_to_host_array_with_sparse_support(arg) for arg in args
+        )
 
         # put all the kwargs on host
         new_kwargs = dict()

python/cuml/cuml/internals/input_utils.py

@@ -530,6 +530,11 @@ def input_to_host_array(
 
 
 def input_to_host_array_with_sparse_support(X):
+    try:
+        if scipy_sparse.isspmatrix(X):
+            return X
+    except UnavailableError:
+        pass
     _array_type, is_sparse = determine_array_type_full(X)
     if is_sparse:
         if _array_type == "cupy":

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

@@ -22,10 +22,17 @@
 from sklearn.decomposition import TruncatedSVD
 from sklearn.kernel_ridge import KernelRidge
 from sklearn.linear_model import (
-    LinearRegression, LogisticRegression, ElasticNet,
-    Ridge, Lasso
+    LinearRegression,
+    LogisticRegression,
+    ElasticNet,
+    Ridge,
+    Lasso,
+)
+from sklearn.neighbors import (
+    NearestNeighbors,
+    KNeighborsClassifier,
+    KNeighborsRegressor,
 )
-from sklearn.neighbors import NearestNeighbors, KNeighborsClassifier, KNeighborsRegressor
 from sklearn.base import is_classifier, is_regressor
 
 
@@ -44,15 +51,14 @@
 
 @pytest.mark.parametrize("estimator_name", list(estimators.keys()))
 def test_sparse_support(estimator_name):
-    X_sparse = csr_matrix([[0, 1],
-                           [1, 0]])
+    X_sparse = csr_matrix([[0, 1], [1, 0]])
     print(X_sparse.shape[0])
     y_class = np.array([0, 1])
     y_reg = np.array([0.0, 1.0])
     estimator = estimators[estimator_name]()
     # Fit or fit_transform depending on the estimator type
     if isinstance(estimator, (KMeans, DBSCAN, TruncatedSVD, NearestNeighbors)):
-        if hasattr(estimator, 'fit_transform'):
+        if hasattr(estimator, "fit_transform"):
             estimator.fit_transform(X_sparse)
         else:
             estimator.fit(X_sparse)