-
-
Notifications
You must be signed in to change notification settings - Fork 50
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
successive SMC inference loops #379
Comments
CC @ciguaran |
Sure, it makes sense to have more than one instance/"chain". Then we would be able to apply diagnostics like r-hat and rankplots. Other diagnostics like ess doesn't make sense. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi there,
I was experimentng with jax sampling and instead of running successive calls to sample_smc_blackjax (which I thought would be useful to compare the posteriors and enable rhat-like diagnostics), I was wondering whether it would make sense to hard code a given number of (successive) jobs in the code itself (pymc_experimental/inference/smc/sampling.py). Does it make sense to build the kernel only once, then runs several inference_loop + arviz_from_particles?
Cheers,
Vian
The text was updated successfully, but these errors were encountered: