embedding density update (#464)

* embedding density

* add release note

* fix

Author: Intron7

docs/release-notes/0.13.3.md

@@ -12,6 +12,7 @@
 
 ```{rubric} Bug fixes
 ```
+* Updates `tl.embedding_density` to work with `rapids-25.10` {pr}`464` {smaller}`S Dicks`
 
 ```{rubric} Misc
 ```

src/rapids_singlecell/tools/_embedding_density.py

@@ -129,9 +129,7 @@ def embedding_density(
         embed_x = adata.obsm[f"X_{basis}"][:, components[0]]
         embed_y = adata.obsm[f"X_{basis}"][:, components[1]]
 
-        adata.obs[density_covariate] = _calc_density(
-            cp.array(embed_x), cp.array(embed_y), batchsize
-        )
+        adata.obs[density_covariate] = _calc_density(embed_x, embed_y, batchsize)
 
     # Reduce diffmap components for labeling
     # Note: plot_scatter takes care of correcting diffmap components
@@ -152,15 +150,15 @@ def _calc_density(x: cp.ndarray, y: cp.ndarray, batchsize: int):
     from cuml.neighbors import KernelDensity
 
     # Calculate the point density
-    xy = cp.vstack([x, y]).T
-    bandwidth = cp.power(xy.shape[0], (-1.0 / (xy.shape[1] + 4)))
+    xy = np.vstack([x, y]).T
+    bandwidth = np.power(xy.shape[0], (-1.0 / (xy.shape[1] + 4)))
     kde = KernelDensity(kernel="gaussian", bandwidth=bandwidth).fit(xy)
     z = cp.zeros(xy.shape[0])
     n_batches = math.ceil(xy.shape[0] / batchsize)
     for batch in range(n_batches):
         start_idx = batch * batchsize
         stop_idx = min(batch * batchsize + batchsize, xy.shape[0])
-        z[start_idx:stop_idx] = kde.score_samples(xy[start_idx:stop_idx, :])
+        z[start_idx:stop_idx] = cp.array(kde.score_samples(xy[start_idx:stop_idx, :]))
     min_z = cp.min(z)
     max_z = cp.max(z)