Hi, I have a model which is a monoexponential decay as a function of some b-values b for which I want to estimate the decay constant ADC. The model as a function of the b_values is like: y_i=s_0\exp\{-b_i\times ADC\} + \epsilon_i where \epsilon is a i.i.d Gaussian noise with standard deviation \…

Try moving the custom log_prob function outside of the pm.Model block “Lambdas defined in local namespace are not picklable (see issue #1995 )” ref: pymc-devs/pymc3/blob/a01d015763f82b590a8355fec018ce2706982607/pymc3/examples/custom_dists.py#L27-L32 # define loglikelihood outside of the m…

Hi, I could find a way for my code to work. Setting the number of cores to 1 cores=1 seems to solve the problem. I don’t know why but I found that most of my models fail if I don’t set the number of cores to 1. Anyone got an idea on the reason why I have such a behaviour? I am on windows and I run …

Thanks, it works! Your help is much appreciated!:slightly_smiling_face:

I have one more question though: How can I pass other parameters that are not random variables or observed data to my function? Thanks!

Question about DensityDist, again

Questions

junpenglao October 23, 2019, 11:41am 6

You can take a look at this post first: How to set up a custom likelihood function for two variables

Topic		Replies	Views
Problem with pymc3.DensityDist() #2846 Questions from_github	3	2323	February 8, 2018
How to do linear fitting with pymc3 by using pm.DensityDist Questions	6	1093	May 4, 2018
How to use costum logliklihood function effectively using pm.DensityDist	8	236	April 18, 2024
pm.DensityDist problem under PMYC 4 v5	17	2618	June 14, 2022
Creating custom likelihoods with DensityDist - what am I doing wrong? Questions	1	884	November 5, 2017

Question about DensityDist, again

Related topics