I’ve just today started trying to use rstan with RStudio. I have been going through the steps as outlined in the “RStan Getting Started.”
RStan Getting Started · stan-dev/rstan Wiki · GitHub.
I’m on a Mac - Ventura. I have restarted R and RStudio multiple times, updated all versions (now R 4.3.1 (2023-06-16) – “Beagle Scouts” and RStudio 2023.06.01+524). I have also updated all packages, and I’ve tried a plethora of troubleshooting options that I have found available. I have not been successful with any.
I am able to install rstan and StanHeaders, and I have gone through the steps for Configuring the C++ Toolchain. On my first attempts, once I got to running
example(stan_model, package = “rstan”, run.dontrun = TRUE)
my RSession aborted every time. After a machine restart and an uninstall/reinstall of rstan, I was able to run the example, but then once I tried to move on to the simple 8 Schools example, I found that RStudio was suppressing errors and was not returning any fit object for
fit ← stan(file = ‘schools.stan’, data = schools_dat)
but it was giving an error only for the first chain that said “Error in unserialize(socklist[[n]]) : error reading from connection.” Unlike other posts that described this issue, it still ran through the other chains but did not return an object.
After some more troubleshooting and restarts, I am now back to being unable to run the
example(stan_model, package = “rstan”, run.dontrun = TRUE)
Specifically, I get the error that says “Error in prep_call_sampler(object) : could not find function “prep_call_sampler””. From my understanding, this can really only happen if rstan is not installed properly, but I’ve tried multiple reinstalls including reinstalls from repo and source all to no avail. If I try the install form repo or source, I also sometimes get an error that StanHeaders is corrupted?
Is there any known fix for this? Any help at all would be appreciated!