Skip to content

Return Euclidean distances from KMeans transform #8539

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Hashim1999164 wants to merge 2 commits into
base: main
Choose a base branch
from
+31 −0
Open

Return Euclidean distances from KMeans transform #8539

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c1afc7b
Return Euclidean distances from KMeans transform
Hashim1999164 Sep 1, 2026
f971d8f
Use cuVS L2SqrtExpanded for KMeans transform distances.
Hashim1999164 Sep 4, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions python/cuml/cuml/cluster/kmeans.pyx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1116,6 +1116,7 @@ class KMeans(
cdef handle_t* handle_ = <handle_t*><size_t>handle.getHandle()
cdef lib.KMeansParams params
_kmeans_init_params(self, params)
params.metric = DistanceType.L2SqrtExpanded

cdef bool values_f32 = X.dtype == cp.float32
cdef bool indices_i32 = _kmeans_indices_i32(n_rows, n_cols)
Expand Down
30 changes: 30 additions & 0 deletions python/cuml/tests/test_kmeans.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -545,6 +545,36 @@ def test_kmeans_device_buffer_samples_host_path(
assert adjusted_rand_score(dev_labels, host_labels) >= 0.97


def test_kmeans_transform_euclidean():
"""transform should return Euclidean distances, matching sklearn."""
X = np.array(
[
[0.0, 0.0],
[1.0, 1.0],
[2.0, 2.0],
[10.0, 10.0],
[11.0, 11.0],
[12.0, 12.0],
]
)
init = np.array([[1.0, 1.0], [11.0, 11.0]])
query = np.array([[0.0, 0.0], [10.0, 10.0]])

cuml_model = cuml.KMeans(
n_clusters=2, init=init, n_init=1, output_type="numpy"
)
cuml_model.fit(X)
cu_dist = cuml_model.transform(query)

sk_model = cluster.KMeans(n_clusters=2, init=init, n_init=1)
sk_model.fit(X)
sk_dist = sk_model.transform(query)

expected = np.sqrt(np.array([[2.0, 242.0], [162.0, 2.0]]))
np.testing.assert_allclose(cu_dist, expected)
np.testing.assert_allclose(cu_dist, sk_dist)


def test_get_feature_names_out():
X, _ = make_blobs(n_features=5)
cu_model = cuml.KMeans().fit(X)
Expand Down
Loading