I am trying to learn 12 parameters over a simplex. The data to fit is a matrix P that is (K, 12) in shape and the observation vector (K, 1)-shaped. I am using this model: with pm.Model() as model: f = pm.Dirichlet("f", np.ones(12)) n_ = pm.Binomial("n", n=2, p=pm.math.dot(P, f), observed=n) i…

Sampling 12 params from 10^5 binomials

ricardoV94 November 23, 2023, 4:49am 4

How many cores do you have?

For why running in parallel may be slower and what you can do about it check out: Frequently Asked Questions - #19 by ricardoV94