Hi all, Following the remark of @jessegrabowski , I checked if .dist nodes are updated. But it looks like they are not, and I am stuck on how to proceed. Specifically, I am motivated by the following problem. Suppose that I want distribute N values according to some distribution with the mean \mu a…

log_norm is the lognormal distribution parameterized by μlog and σlog. It’s a latent variable in your model. I looked at things more closely, and there are several improvements you can make: You can dispense with the scans entirely. You are merely iterating over a count variable, so you can accom…

Difficulties in using .dist and pm.Potential for inference

Questions v5

cluhmann October 15, 2022, 8:53pm 3

I suspect that the various pieces connecting \mu and \sigma to the likelihood are not being sampled because they are not registered in the model. @ricardoV94?

Topic		Replies	Views
DensityDist from scipy.stats distribution Questions	5	2287	May 20, 2018
Mu of pm.Normal() depends on output of RandomVariable on previous time step. How do I implement this? v5 time_series , modeling	8	771	July 26, 2023
pm.Deterministic remains unchanged although the parameters are changing v5	5	496	October 14, 2022
Declare an internal Poisson distribution inside a custom distribution failes Questions	21	2096	April 30, 2018
Rewriting Likelihood with DensityDist() outputs MissingInputError Questions	9	848	April 3, 2021

Difficulties in using .dist and pm.Potential for inference

Related topics