Error with Bayesian workflow for disease transmission tutorial

Modeling
, ,
1

I’m trying to follow along with Bayesian workflow for disease transmission modeling in Stan, but unfortunately I can’t get the example to run. I’ve used exactly the code provided in the tutorial with no alterations, but I get the error

[1] "Error in sampler$call_sampler(args_list[[i]]) : "
[2] "  c++ exception (unknown reason)"                
error occurred during calling the sampler; sampling not done

I found this, which seemed to suggest the error could be within the negative binomial function. But the suggested fix doesn’t appear to work for me. Any thoughts?

2

Hi Ian. I’m unfamiliar with rStan but just by your description alone I assume the problem arises when you’re trying to compile the Stan model. Have you come across this by any chance:

Also, just to confirm, did you save the model in “sir_negbin.stan” as suggested in the source and then had the problem when you run the lines below?

model <- stan_model("sir_negbin.stan")
fit_sir_negbin <- sampling(model,
                data = data_sir,
                iter = niter,
                chains = 4)
1 Like
3

@charlesm93

4

Yep, that’s right where I have the problem. I’ve noticed if I comment out the line

cases ~ neg_binomial_2(col(to_matrix(y), 2), phi);

within the “sir_negbin.stan” file I don’t have an issue

5

I’ve solved the issue, in case anybody runs into this in the future.
I can’t believe I didn’t try this earlier, but I hadn’t installed the latest release of R (version 4.0.2 ). I did that and reinstalled rstan, and now everything is working.

2 Likes