mcmc_pairs() works differently on an HPC cluster?

[JasonPekos]

I can create pairs plots fine on my laptop.

mcmc_pairs(mcmc_out$draws(), np = nuts_params(mcmc_out))

When I run the identical code on the Sharcnet HPC system, if I include np = nuts_params(mcmc_out), the code will fail with the error:

Error in `geom_point()`:
! Problem while computing aesthetics.
ℹ Error occurred in the 2nd layer.
Caused by error in `check_aesthetics()`:
! Aesthetics must be either length 1 or the same as the data (15000)
✖ Fix the following mappings: `colour` and `size`
Run `rlang::last_error()` to see where the error occurred.

[ASKurz]

I'm running into this problem with brm() models (brms 2.18.0; bayesplot 1.10.0).

[rewberl]

Also experiencing this issue with a model fit using cmdstanr (2.35.0)

[jgabry]

Thanks for bringing this up again. Sorry it's still an issue. Unfortunately I was never able to reproduce this on the cluster I had access to, which made it quite difficult to debug. If anyone has any ideas what might be causing this I'd love to know so we can fix it!