stanc: Translate Stan model specification to C++ code

View source: R/stanc.R

stancR Documentation

Translate Stan model specification to C++ code


Translate a model specification in Stan code to C++ code, which can then be compiled and loaded for sampling.


  stanc(file, model_code = '', model_name = "anon_model", verbose = FALSE,
        obfuscate_model_name = TRUE, allow_undefined = FALSE,
        isystem = c(if (!missing(file)) dirname(file), getwd())) 
  stanc_builder(file, isystem = c(dirname(file), getwd()),
                verbose = FALSE, obfuscate_model_name = FALSE,
                allow_undefined = FALSE)



A character string or a connection that R supports specifying the Stan model specification in Stan's modeling language.


Either a character string containing a Stan model specification or the name of a character string object in the workspace. This parameter is used only if parameter file is not specified, so it defaults to the empty string.


A character string naming the model. The default is "anon_model". However, the model name will be derived from file or model_code (if model_code is the name of a character string object) if model_name is not specified.


Logical, defaulting to FALSE. If TRUE more intermediate information is printed during the translation procedure.


Logical, defaulting to TRUE, indicating whether to use a randomly-generated character string for the name of the C++ class. This prevents name clashes when compiling multiple models in the same R session.


A character vector naming a path to look for file paths in file that are to be included within the Stan program named by file. See the Details section below.


A logical scalar defaulting to FALSE indicating whether to allow Stan functions to be declared but not defined in file or model_code. If TRUE, then it is the caller's responsibility to provide a function definition in another header file or linked shared object.


The stanc_builder function supports the standard C++ convention of specifying something like #include "my_includes.txt" on an entire line within the file named by the file argument. In other words, stanc_builder would look for "my_includes.txt" in (or under) the directories named by the isystem argument and — if found — insert its contents verbatim at that position before calling stanc on the resulting model_code. This mechanism reduces the need to copy common chunks of code across Stan programs. It is possible to include such files recursively.

Note that line numbers referred to in parser warnings or errors refer to the postprocessed Stan program rather than file. In the case of a parser error, the postprocessed Stan program will be printed after the error message. Line numbers referred to in messages while Stan is executing also refer to the postprocessed Stan program which can be obtained by calling get_stancode.


A list with named entries:

  1. model_name Character string for the model name.

  2. model_code Character string for the model's Stan specification.

  3. cppcode Character string for the model's C++ code.

  4. status Logical indicating success/failure (always TRUE) of translating the Stan code.


Unlike R, in which variable identifiers may contain dots (e.g. a.1), Stan prohibits dots from occurring in variable identifiers. Furthermore, C++ reserved words and Stan reserved words may not be used for variable names; see the Stan User's Guide for a complete list.


The Stan Development Team Stan Modeling Language User's Guide and Reference Manual.

The Stan Development Team CmdStan Interface User's Guide.

See Also

stan_model and stan


stanmodelcode <- "
data {
  int<lower=0> N;
  real y[N];

parameters {
  real mu;

model {
  mu ~ normal(0, 10);
  y ~ normal(mu, 1); 

r <- stanc(model_code = stanmodelcode, model_name = "normal1") 

rstan documentation built on Sept. 8, 2022, 5:06 p.m.