scib_metrics.utils.cdist

scib_metrics.utils.cdist(x, y, metric='euclidean')

Jax implementation of scipy.spatial.distance.cdist().

Uses euclidean distance by default, cosine distance is also available.

Parameters:

x ndarray

Array of shape (n_cells_a, n_features)

y ndarray

Array of shape (n_cells_b, n_features)

metric Literal['euclidean', 'cosine'] (default: 'euclidean')

The distance metric to use. The distance function can be ‘euclidean’ (default) or ‘cosine’.

Return type:

Array

Returns:

dist Array of shape (n_cells_a, n_cells_b)