I am using a very complex function as likelihood, that makes extensive use of true_div, real, sqrt, etc. I am on a Linux machine in a dedicated conda environment, PyMC5 with jax support, everything installed with conda. What I am missing?

Do you also see that if you set cores=1 in sample?

The Op for true_div, real, sqrt and others does not have a C implementation

verderis March 6, 2023, 9:42pm 3

I am using version 5.0.2

Topic		Replies	Views
@lucianopaz。Please help me solve the Windows multiprocessing issue Questions	33	3232	December 8, 2018
Problem with multiprocessing in PyMC3 Questions	5	3722	August 20, 2018
ERROR (theano.gof.opt) & slowness in execution in sampling Questions	3	1168	March 27, 2019
Sampling issue: The following error happened while compiling the node Questions theano	4	3294	January 21, 2019
WARNING (theano.tensor.blas): Using NumPy C-API based implementation for BLAS functions.——–>HELP! Questions	2	3808	January 24, 2020