I would like to build the likelihood on several matrices simultaneously, i.e., my observed data (tensor) is of shape (Z, M, N)=(885, 50, 6). K = 5 T = 50 P = 6 N = data_x.shape[0] with pm.Model(coords={"component": np.arange(K), "obs_id": np.arange(N)}) as model: # DP prior pi = pm.StickBr…

How to use MatrixNormal for modeling several matrices?

ricardoV94 January 18, 2024, 7:55am 2

Your dims don’t match the shape of your observed data. You should have as many dims as data dimensions.

Topic		Replies	Views
Model multivariate normal with separate means, dimension mismatch error Questions	2	3090	August 7, 2017
Generating a random matrix with MvNormal Questions	4	745	December 4, 2019
About set of Multivariate Normal Distributions Questions	2	1187	July 3, 2018
2D Gaussian Mixture Model Questions	7	2095	August 25, 2017
Draw number of MvNormal in once Questions	2	575	April 5, 2019