C++ exception rstan error, Stan model does not contain samples. Catalina 10.15.3, rstan 2.21.2

Interfaces RStan
1

Hi there! I have been trying to use rstan to run my bayesian model, but have been receiving a c++ exception error message and “Stan model does not contain samples”. I am using rstan version 2.21.2, R-3.6.3 on Mac Catalina 10.15.3. Here is some code for the model I’m running, as well as the subsequent error messages. Any help at all would be much appreciated!!

bayes_name <- mm_name(type='bayes', pool_K600='binned',
                      err_obs_iid=TRUE, err_proc_acor=FALSE, err_proc_iid=TRUE, 
                      ode_method = 'trapezoid', deficit_src='DO_mod', engine='stan')

#### Set the model specs. 
bayes_specs <- revise(specs(bayes_name), 
                      burnin_steps=1000, saved_steps=500,
                      day_start=3, day_end=27,
                      GPP_daily_mu = 1, GPP_daily_lower = 0,GPP_daily_sigma = 2,
                      ER_daily_mu = -6, ER_daily_upper = 0, ER_daily_sigma = 3,
                      K600_lnQ_nodes_centers= log(pred_K$discharge),
                      K600_lnQ_nodes_meanlog= log(pred_K$K.600.fit),
                      K600_daily_sigma_sigma=0.7, chains=1) #
bayes_specs

#run the model with the specs
bayes_fit_kup <- metab(bayes_specs, data=dat_bayes)

## ERROR MESSAGES:
Chain 1: Iteration:    1 / 1500 [  0%]  (Warmup)
[1] "Error in sampler$call_sampler(args_list[[i]]) : "
[2] "  c++ exception (unknown reason)"                
error occurred during calling the sampler; sampling not done
here are whatever error messages were returned
`data_frame()` is deprecated as of tibble 1.1.0.
Please use `tibble()` instead.
This warning is displayed once every 8 hours.
Call `lifecycle::last_warnings()` to see where this warning was generated.Modeling failed
  Warnings:
    some chains had errors; consider specifying chains = 1 to debug
    Stan model 'b_Kb_oipi_tr_plrckm' does not contain samples.
2

Getting Stan stuff to work with R 3.x on Macs is not impossible but it is very difficult. In contrast, it is fairly easy if you upgrade to R 4.x, in which case you basically just need Xcode and to delete ~/.R/Makevars and ~/.Renviron . Is there a strong reason why you cannot upgrade to the latest R?

3

Hi there! Thank you for your response. I was concerned with the compatibility of my current code with R 4.x but if it’s a generally seamless transition I can give it a shot. Do R upgrades tend to require alterations to code written in older versions?

4

No

5

Thank you for your help. I just updated to R 4.0.3 and am now having trouble even installing stan. The Rcpp, Rcpparmadillo, and xcode toolchains have all been downloaded, but I keep getting an error that reads as below. Any recommendations on installing on R 4.0.3 Catalina mac?

"clang: error: no such file or directory: '/Library/Frameworks/R.framework/Resources/lib/libc++abi.1.dylib'
make: *** [rstan.so] Error 1
ERROR: compilation failed for package ‘rstan’
* removing ‘/Library/Frameworks/R.framework/Versions/4.0/Resources/library/rstan’

when i use the call:

install.packages("rstan", repos = "https://cloud.r-project.org/", dependencies = TRUE)
6

Delete the ~/.R/Makevars and ~/.Renviron files.

7

I deleted the environment file but didn’t “have permission” to delete the makevars. According to earlier inquiries, my makevars code appears correct:

# clang: start
CFLAGS=-isysroot /Library/Developer/CommandLineTools/SDKs/MacOSX.sdk
CCFLAGS=-isysroot /Library/Developer/CommandLineTools/SDKs/MacOSX.sdk
CXXFLAGS=-isysroot /Library/Developer/CommandLineTools/SDKs/MacOSX.sdk
CPPFLAGS=-isysroot /Library/Developer/CommandLineTools/SDKs/MacOSX.sdk -I/usr/local/include

SHLIB_CXXLDFLAGS+=-Wl,-rpath,${R_HOME}/lib ${R_HOME}/lib/libc++abi.1.dylib
SHLIB_CXX14LDFLAGS+=-Wl,-rpath,${R_HOME}/lib ${R_HOME}/lib/libc++abi.1.dylib
# clang: end

if this is incorrect should i just delete the code in the makevars file and proceed with attempting to install rstan?

8

You may have to use sudo to delete the file, but you shouldn’t have any of that to use rstan with R 4.x.