I’m new to pyMC3 and am still struggling to make sense of the API, with that said, my issue appears to be a dependency issue with multiprocessing I’m attempting to make a mixture model, specifying each distribution separately and then assigning them to comp_dists kwarg of Mixture() as a list. See …

AttributeError: Can't pickle local object 'Mixture._comp_dist_random_wrapper.<locals>.wrapped_random'

jstanley March 6, 2020, 5:56pm 5

Setting the number of cores equal to 1 solved the issue. Specifically:

trace1 = pm.sample(1000, cores=1)

I’ll raise a bug on Github.

1 Like

Topic		Replies	Views
AttributeError: Can't pickle local object 'Op.make_py_thunk..rval' while running pymc3 Questions	1	986	June 29, 2018
Problem with multiprocessing in PyMC3 Questions	5	3722	August 20, 2018
KeyError Setting Mixture Proportions for Mixture Model Questions	26	1503	May 13, 2019
Multiprocess sampling without using Pickle Questions	2	1499	June 16, 2018
Is there an example on how to work with generalized mixture models? Questions	15	3593	March 16, 2019