class bayespy.inference.vmp.transformations.RotateVaryingMarkovChain(X, B, S, B_rotator)[source]

Rotation for bayespy.nodes.SwitchingGaussianMarkovChain

Assume the following model.

Constant, unit isotropic innovation noise.

A_n = \sum_k B_k s_{kn}

Gaussian B: (1,D) x (D,K) Gaussian S: (N,1) x (K) MC X: () x (N+1,D)

No plates for X.

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

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


__init__(X, B, S, 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.