state.priors#

Prior distributions for state model parameters.

Classes

LatentsPriorStatic

Static latent prior: X ~ N(0, I).
class LatentsPriorStatic[source]#

Static latent prior: X ~ N(0, I).

The standard GFA prior assumes independent standard normal latents.

Examples

>>> prior = LatentsPriorStatic()
>>> rng = np.random.default_rng(42)
>>> X = prior.sample(x_dim=5, n_samples=100, rng=rng)
>>> X.data.shape
(5, 100)
sample(
x_dim: int,
n_samples: int,
rng: Generator,
) → LatentsRealization[source]#

Sample X ~ N(0, I).

Parameters:
x_dimint

Number of latent dimensions.

n_samplesint

Number of samples to generate.

rngnumpy.random.Generator

Random number generator.

Returns:
LatentsRealization

Sampled latent values.