Hi, I’m toying with a Normal MixtureModel. The weights are Dirichlet: import pymc3 as pm import numpy as np import theano.tensor as tt W = np.array([0.35, 0.4, 0.25]) with pm.Model() as model: w = pm.Dirichlet('w', np.ones_like(W)) mu = pm.Normal('mu', 0., 10., shape=W.size) tau = pm.Gamma('ta…

Dirichlet with random shape

junpenglao February 14, 2023, 7:01am 7

We are definitely better at supporting dynamic shape, but I am not sure if your use case is supported yet. Could you open a new post with the example you have in mind?

Topic		Replies	Views
Dynamic number of Dirichlet distributions? Questions	3	622	February 25, 2018
Discrete hyper-parameter input for Dirichlet prior Questions	1	423	March 18, 2020
Shape issues when creating Dirichlet mixtures Questions	0	434	August 19, 2021
Shape error when sampling more than one value Questions	13	1907	December 20, 2018
Weird shape behavior/error using a mixture Questions	0	416	March 2, 2020

Dirichlet with random shape

Related topics