Any suggestions on continous error "ValueError: shape mismatch: value array of shape (62,) could not be broadcast to indexing result of shape (0,)"

iavicenna · March 28, 2024, 9:59am

I have seen in some instances that supplying chol instead of cov speeds up mvnormal sampling. And that seems to be the recommended usage:

Also have you tried NUTs instead of Metropolis? Finally you can also try numpyro:

Topic		Replies	Views
Shape error when sampling more than one value Questions	13	1907	December 20, 2018
KeyError Setting Mixture Proportions for Mixture Model Questions	26	1503	May 13, 2019
Sampling issue: The following error happened while compiling the node Questions theano	4	3294	January 21, 2019
How to deal with a variable size of the 'observed' variable Questions	15	4003	October 16, 2017
ERROR (theano.gof.opt) & slowness in execution in sampling Questions	3	1168	March 27, 2019