suppress all output from RStan

[bob-carpenter]

Requested by Avraham Adler on stan-users:

Firstly, as a medium-term lurker but first-time poster, I would like to thank the entire Stan development team for the creation and maintenance of Stan.

I am using Stan as part of a paper I am trying to write, and I am struggling with trying to suppress diagnostic messages. For example, when I have used JAGS, R2jags, and knitr, once I set echo, messages, and warnings to FALSE, I can have the model run every time I knit the paper, and only have the results I want displayed through \Sexpr{}. Using Stan, I am having much more of a problem. I have searched the group and have found the following threads:

• Suppressing Warning Messages
• silent stan?
• RStan print() - Trivial request

I have verbose set to FALSE and refresh set to -1. I have used suppressMessages/Warnings as suggested in the thread (1), and that does not seem to do anything more than setting messages and warnings settings in the knitr chunk. I have wrapped my stan() call in sink() calls, as discussed in thread (2) as such:

<<'Bayes_Stan', eval=TRUE, echo=FALSE, message=FALSE, warning=FALSE, cache=0>>=
sink(file = "a", type = c("output", "message"))
library(rstan)
LN_S <- suppressMessages(suppressWarnings(stan(file = 'Bayes.stan', data = BayesData, pars = c('mu', 'sigma', 'LR_post', 'ASL_post', 'dev'), iter = 300000, warmup = 50000, thin = 25, chains = 5, seed = 12, refresh = -1, verbose = FALSE)))

[snip extractions and assignments]

LNPlus_S <- suppressMessages(suppressWarnings(stan(file = 'BayesPlus.stan', data = BayesDataPlus, pars = c('Pars', 'LR_post', 'ASL_post', 'dev'), iter = 300000, warmup = 50000, thin = 25, chains = 5, seed = 12, refresh = -1, verbose = FALSE)))

[snip extractions and assignments]
sink()
@

Regardless, my knitted output looks like:

…Therefore, the model was rebuilt in Stan (Stan Development Team, 2014a)—a Bayesian modeling language whose samplers are non-Gibbs.
##
## TRANSLATING MODEL 'Bayes' FROM Stan CODE TO C++ CODE NOW.
## COMPILING THE C++ CODE FOR MODEL 'Bayes' NOW.
## cygwin warning:
## MS-DOS style path detected: C:/R/RCurrent/R-30~1.2/etc/x64/Makeconf
## Preferred POSIX equivalent is: /cygdrive/c/R/RCurrent/R-30~1.2/etc/x64/Makeconf
## CYGWIN environment variable option "nodosfilewarning" turns off this warning.
## Consult the user's guide for more details about POSIX paths:
## http://cygwin.com/cygwin-ug-net/using.html#using-pathnames
## C:/R/RCurrent/R-3.0.2/library/rstan/include//stansrc/stan/agrad/rev/var_stack.hpp:49:17: warning: 'void stan::agrad::free_memory()' defined but not used [-Wunused-function]
## C:/R/RCurrent/R-3.0.2/library/rstan/include//stansrc/stan/agrad/rev/chainable.hpp:87:17: warning: 'void stan::agrad::set_zero_all_adjoints()' defined but not used [-Wunused-function]
##
## TRANSLATING MODEL 'BayesPlus' FROM Stan CODE TO C++ CODE NOW.
## COMPILING THE C++ CODE FOR MODEL 'BayesPlus' NOW.
## cygwin warning:
## MS-DOS style path detected: C:/R/RCurrent/R-30~1.2/etc/x64/Makeconf
## Preferred POSIX equivalent is: /cygdrive/c/R/RCurrent/R-30~1.2/etc/x64/Makeconf
## CYGWIN environment variable option "nodosfilewarning" turns off this warning.
## Consult the user's guide for more details about POSIX paths:
## http://cygwin.com/cygwin-ug-net/using.html#using-pathnames
## C:/R/RCurrent/R-3.0.2/library/rstan/include//stansrc/stan/agrad/rev/var_stack.hpp:49:17: warning: 'void stan::agrad::free_memory()' defined but not used [-Wunused-function]
## C:/R/RCurrent/R-3.0.2/library/rstan/include//stansrc/stan/agrad/rev/chainable.hpp:87:17: warning: 'void stan::agrad::set_zero_all_adjoints()' defined but not used [-Wunused-function]
## Error in function stan::prob::lognormal_log(N4stan5agrad3varE): Scale parameter is -0.34910650273000776:0, but must be > 0!Rejecting proposed initial value with zero density.
## Error in function stan::prob::lognormal_log(N4stan5agrad3varE): Scale parameter is -1.6172265406146098:0, but must be > 0!Rejecting proposed initial value with zero density.
## Error in function stan::prob::lognormal_log(N4stan5agrad3varE): Scale parameter is -1.2449912107872052:0, but must be > 0!Rejecting proposed initial value with zero density.
## Error in function stan::prob::lognormal_log(N4stan5agrad3varE): Scale parameter is -1.0263658781850085:0, but must be > 0!Rejecting proposed initial value with zero density.
## Error in function stan::prob::lognormal_log(N4stan5agrad3varE): Scale parameter is -1.7036664255500364:0, but must be > 0!Rejecting proposed initial value with zero density.
## Error in function stan::prob::lognormal_log(N4stan5agrad3varE): Scale parameter is -1.7912999475616009:0, but must be > 0!Rejecting proposed initial value with zero density.
## Error in function stan::prob::lognormal_log(N4stan5agrad3varE): Scale parameter is -0.67853001987262718:0, but must be > 0!Rejecting proposed initial value with zero density.
## Error in function stan::prob::lognormal_log(N4stan5agrad3varE): Scale parameter is -0.56723437122169695:0, but must be > 0!Rejecting proposed initial value with zero density.
## Error in function stan::prob::lognormal_log(N4stan5agrad3varE): Scale parameter is -1.9622578028376079:0, but must be > 0!Rejecting proposed initial value with zero density.
## Error in function stan::prob::lognormal_log(N4stan5agrad3varE): Scale parameter is -1.9058489407892381:0, but must be > 0!Rejecting proposed initial value with zero density.
The No-U-Turn-Sampler behind Stan is much more robust to highly autocorrelated parameters…

While I can always knit the Rnw, edit the resulting tex file to remove these lines, and re-texify that file, I would appreciate knowing if there is any way to prevent the need for such post-processing.

[aadler]

Thank you for posting this, Bob.

--Avraham

[maverickg]

Those error messages are from Stan's c++ code. Which one of standard output and standard error output (std::cout or std::cerr) is used for them?

[syclik]

@maverickg, it's now piped to Rcpp::Rcout. Let me find the lines that we control that.

[syclik]

@aadler, do you have a simple example script that I can try? I have the latest and greatest RStan installed and could see if the refactoring fixed this issue.

[aadler]

Let me see what I can extract from my .Rnw. I'll send it to you offline once I have something.

[ksvanhorn]

The optimizing() function in RStan also has this issue -- even with verbose=FALSE, it still prints out a stream of progress messages and parameter settings. I have an application in which I typically run around 1000 calls to optimizing(), each of which takes a few seconds; the voluminous output this produces is a real annoyance.

[aadler]

Personally, I've resorted to just kniting the .Rnw and then open the resulting .tex file in Winedt and deleting all those statements. As things go, it's a minor annoyance which I can live with, although I certainly would appreciate damping them.

Dr. Denwood of runjags was able to dampen all output after I asked a while ago, maybe he has an idea? Then again, runjags is MUCH simpler than stan and lives completely within R, so there probably isn't much comparison.

Avi

[bob-carpenter]

We'll get this fixed soon. We're taking it on
with our general infrastructure refactoring to try to
share more code among CmdStan, RStan, and PyStan (MatlabStan
and Stan.jl just call CmdStan).

The problem is not in RStan, but in the C++ API for Stan itself.

• Bob

On Jan 22, 2015, at 7:11 PM, Avraham Adler notifications@github.com wrote:

Personally, I've resorted to just kniting the .Rnw and then open the resulting .tex file in Winedt and deleting all those statements. As things go, it's a minor annoyance which I can live with, although I certainly would appreciate damping them.

Dr. Denwood of runjags was able to dampen all output after I asked a while ago, maybe he has an idea? Then again, runjags is MUCH simpler than stan and lives completely within R, so there probably isn't much comparison.

Avi

—
Reply to this email directly or view it on GitHub.

[alyst]

PR #154 partly addresses the issue.

[syclik]

this is my reminder to fix this after the refactor.

[camlionur]

hi,
I dont want to see details of rstan output while rstan is working. so how can I suppress the Rstan output as seen below? thanks.

AMPLING FOR MODEL 'test_par_est' NOW (CHAIN 1).

Chain 1, Iteration: 1 / 10000 0%
Chain 1, Iteration: 1000 / 10000 10%
Chain 1, Iteration: 2000 / 10000 20%
Chain 1, Iteration: 3000 / 10000 30%
Chain 1, Iteration: 4000 / 10000 40%
Chain 1, Iteration: 5000 / 10000 50%
Chain 1, Iteration: 5001 / 10000 50%
Chain 1, Iteration: 6000 / 10000 60%
Chain 1, Iteration: 7000 / 10000 70%
Chain 1, Iteration: 8000 / 10000 80%
Chain 1, Iteration: 9000 / 10000 90%
Chain 1, Iteration: 10000 / 10000 100%#

Camlionur