class bayespy.inference.vmp.transformations.RotateSwitchingMarkovChain(X, B, Z, B_rotator)[source]

Rotation for bayespy.nodes.VaryingGaussianMarkovChain

Assume the following model.

Constant, unit isotropic innovation noise.

A_n = B_{z_n}

Gaussian B: (…, K, D) x (D) Categorical Z: (…, N-1) x (K) GaussianMarkovChain X: (…) x (N,D)

No plates for X.

__init__(X, B, Z, B_rotator)[source]

Initialize self. See help(type(self)) for accurate signature.


__init__(X, B, Z, B_rotator) Initialize self.
bound(R[, logdet, inv])
get_bound_terms(R[, logdet, inv])
rotate(R[, inv, logdet])
setup() This method should be called just before optimization.