ott.geometry.semidiscrete_pointcloud.SemidiscretePointCloud.sample

SemidiscretePointCloud.sample(rng, num_samples, *, epsilon=None)

Sample a point cloud.

Note

When is_entropy_regularized = False, some methods and attributes of the sampled PointCloud are not meaningful. However, this does not impact the usage of the SemidiscreteSolver.

Parameters:

• rng (Array) – Random key used for seeding.

• num_samples (int) – Number of samples.

• epsilon (Optional[float]) – Epsilon regularization. If None, use epsilon.

Return type:

PointCloud

Returns:

The sampled point cloud.