spey.backends.distributions.MultivariateNormal

class spey.backends.distributions.MultivariateNormal(mean: ndarray, cov: ndarray, weight: Callable[[ndarray], float] | float = 1.0, domain: slice = slice(None, None, None))

Multivariate normal distribution

Parameters:

• mean (np.ndarray) – Mean of the distribution.

• cov (np.ndarray) – Symmetric positive (semi)definite covariance matrix of the distribution.

• weight (Callable[[np.ndarray], float] or float, default 1.0) – weight of the distribution.

• domain (slice, default slice(None, None)) – set of parameters to be used within the distribution.

__init__(mean: ndarray, cov: ndarray, weight: Callable[[ndarray], float] | float = 1.0, domain: slice = slice(None, None, None))

Methods

__init__(mean, cov[, weight, domain])
expected_data()	The expectation value of the Multivariate Normal distribution.
log_prob(value)	Compute log-probability
sample(value, sample_size)	Generate samples

Attributes

mean	Mean of the distribution.
cov	Symmetric positive (semi)definite covariance matrix of the distribution.
weight	Weight of the distribution
domain	Which parameters should be used during the computation of the pdf