Hi, I’m using pymc5.5.0. I have successfully (I believe!) created a CustomDist: def binom_logp( value, n, p): value = pt.as_tensor_variable(intX(value)) ## do I need these? n = pt.as_tensor_variable(intX(n), dtype='int32') ## do I need these? p = pt.as_tensor_variable(float…

You can pass dtype="int64" to your CustomDist to make it discrete

Making a discrete CustomDist

cluhmann June 13, 2023, 1:54am 2

Can we see the full model (the one you are sampling)?

Topic		Replies	Views
Discrete DensityDist v5 modeling	5	529	January 7, 2023
Recommended way to create a new discrete distribution? Questions development	3	1818	February 4, 2020
Issue with DiscreteUniform distribution Questions	3	961	January 30, 2018
pm.find_MAP won't follow the bounds of custom distribution Questions	1	442	December 9, 2019
Issue with categorical.dist Questions	0	472	May 27, 2019