Hi! I created a custom distribution using a custom likelihood written with pytensor functions only (detailed likelihood implementation is the logp_ddm function here ). I built a basic model with PyMC using this distribution and tried to perform some basic sampling like this: with pm.Model() as ddm_…

"<<! BUG IN FGRAPH.REPLACE OR A LISTENER !>>" during mcmc sampling, but sampling finishes correctly

ricardoV94 July 5, 2023, 1:06pm 4

PyTensor, but it would be good if you could simplify it quite a bit and make it a standalone reproducible snippet

Topic		Replies	Views
Default Type in pm.Mixture Questions	1	452	April 26, 2018
Switch point Metropolis tuning Questions	25	5000	May 19, 2018
Combining parameters into single tensor Questions	19	3538	October 30, 2017
Sampling issue: The following error happened while compiling the node Questions theano	4	3294	January 21, 2019
pm.Potential without Theano? Questions	14	2301	March 20, 2018