Open JasonPekos opened 2 months ago
FWIW I think this can happen in Stan too, but it's super rare, so they're doing something different. If possible, we could force both Stan and Turing to initialize at the same parameter set, and then watch the differences in behaviour.
This is a recurring problem people run into when doing Bayesian ODEs (I think this can happen in Stan too, not just Turing).
Relevant discourse thread: https://discourse.julialang.org/t/turing-nuts-chains-getting-stuck-at-the-parameter-bounds/85947.
My vague intuition for what's going on is:
(with an eval at -Inf)
based on the solver retcode.-Inf
.This might not be right — I'd want to run some tests to confirm this. Frustratingly, this makes benchmarking with
@btime
difficult, because benchmark performance is dominated by stuck chains.I guess solutions would be:
Example of this happening with the LK model: