I am currently setting up a Hierarchical Bayesian Model for Calibration of a Material Model. When I setup my model I define a customised likelihood. When defining the line pm.Potential(“likelihood”, logl(param)) pymc3 appears to test my customised likelihood, yet it does it with parameters a long…

Issue with setting up pm.Potential("likelihood", logl(param))

ricardoV94 January 11, 2021, 6:34am 4

At a first glance it seems like you have a positive log likelihood for 2 of your parameters. That looks wrong, but maybe it’s just unnormalized PDFs?

Topic		Replies	Views
pm.Potential() much needed explanation for newbie Questions	4	8459	December 6, 2018
Problem model definition PyMC3 Questions	12	4749	September 21, 2017
How to set up a custom likelihood function for two variables Questions	8	13306	March 9, 2018
Sampling crashes with "The derivative of RV [whatever] is zero" when using a custom likelihood defined with pm.Potential Questions	4	4028	February 23, 2019
Problem with pymc3.DensityDist() #2846 Questions from_github	3	2284	February 8, 2018