bayespy.inference.vmp.nodes.multinomial.MultinomialDistribution¶
-
class
bayespy.inference.vmp.nodes.multinomial.
MultinomialDistribution
(trials)[source]¶ Class for the VMP formulas of multinomial variables.
-
__init__
(trials)[source]¶ Create VMP formula node for a multinomial variable
trials is the total number of trials.
Methods
__init__
(trials)Create VMP formula node for a multinomial variable compute_cgf_from_parents
(u_p)Compute compute_fixed_moments_and_f
(x[, mask])Compute the moments and for a fixed value.
compute_gradient
(g, u, phi)Compute the Euclidean gradient. compute_logpdf
(u, phi, g, f, ndims)Compute E[log p(X)] given E[u], E[phi], E[g] and E[f]. compute_message_to_parent
(parent, index, u, u_p)Compute the message to a parent node. compute_moments_and_cgf
(phi[, mask])Compute the moments and .
compute_phi_from_parents
(u_p[, mask])Compute the natural parameter vector given parent moments. compute_weights_to_parent
(index, weights)Maps the mask to the plates of a parent. plates_from_parent
(index, plates)Resolve the plate mapping from a parent. plates_to_parent
(index, plates)Resolves the plate mapping to a parent. random
(*phi[, plates])Draw a random sample from the distribution. -