Re: [stan] Problem installing rstan (#1613)

[Bob Carpenter]

Could one of the RStan devs verify? Is there a workaround in place?

The problem the user had who came to our office yesterday was primarily
with dependencies=TRUE. I don't know enough about RStan installs to do much more
than guess.

- Bob

P.S. Here's what Alexandre originally sent:


Hello,

I have been using RStan for nearly a month now and today I tried to use the command "stan_model" and it didn't work, even on the simple example:

stan_model(model_code = 'parameters {real y;} model {y ~ normal(0,1);}')

Therefore I decided to re-install RStan and this is when all my problems started. Here is the way I did it and the messages I got:

.libPaths()
[1] "C:/Users/MOESCHIA/Programs/R/R-3.2.2/library"

Which is correct.

Sys.getenv("PATH")
[1] "C:\Users\MOESCHIA\Programs\R\R-3.2.2\bin\x64;C:\Users\MOESCHIA\Programs\Rtools\bin;C:\Users\MOESCHIA\Programs\Rtools\gcc-4.6.3\bin;C:\Program Files (x86)\Intel\iCLS Client\;C:\Program Files\Intel\iCLS Client\;C:\Program Files (x86)\Java\jre6\bin;C:\Program Files\Java\jre6\bin;C:\Windows\System32;C:\Windows;C:\Windows\System32\wbem;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Program Files\Intel\Intel(R) Management Engine Components\DAL;C:\Program Files\Intel\Intel(R) Management Engine Components\IPT;C:\Program Files (x86)\Intel\Intel(R) Management Engine Components\DAL;C:\Program Files (x86)\Intel\Intel(R) Management Engine Components\IPT;C:\texlive\2015\bin\win32;C:\Program Files (x86)\WebEx\Productivity Tools;C:\Program Files (x86)\WinSCP\;C:\Program Files\MiKTeX 2.9\miktex\bin\x64\;C:\Users\MOESCHIA\Dropbox\Stage\R\RStan\Tests\ C:\Users\MOESCHIA\Programs\Putty"

Which agrees with what is written on the RStan Getting Started guide. Same for:

system('where make')
C:\Users\MOESCHIA\Programs\Rtools\bin\make.exe

Sys.getenv("HOME")
[1] "C:"

Is correct.

dotR <- file.path(Sys.getenv("HOME"), ".R")
if (!file.exists(dotR)) dir.create(dotR)
M <- file.path(dotR, "Makevars")
if (!file.exists(M)) file.create(M)
cat("\nCXXFLAGS=-O3", file = M, sep = "\n", append = TRUE)

This correctly created the file in C:/.R

Sys.setenv(MAKEFLAGS = "-j4")
install.packages("rstan", dependencies = TRUE)
trying URL 'https://cran.rstudio.com/bin/windows/contrib/3.2/rstan_2.7.0-1.zip'
Content type 'application/zip' length 3394676 bytes (3.2 MB)
downloaded 3.2 MB

package ‘rstan’ successfully unpacked and MD5 sums checked

The downloaded binary packages are in
C:\Users\MOESCHIA\AppData\Local\Temp\RtmpWG4l1q\downloaded_packages

Everything is ok too.

library(rstan)
Le chargement a nécessité le package : Rcpp
Le chargement a nécessité le package : inline

Attachement du package : ‘inline’

The following object is masked from ‘package:Rcpp’:

registerPlugin

rstan (Version 2.7.0-1, packaged: 2015-07-17 18:12:01 UTC, GitRev: 05c3d0058b6a)
For execution on a local, multicore CPU with excess RAM we recommend calling
rstan_options(auto_write = TRUE)
options(mc.cores = parallel::detectCores())

Then I tried to run the example in the RStan Getting Started guide.

setwd("C:/Users/MOESCHIA/Dropbox/Stage/R/RStan/Tests") # 8schools.stan was correctly copied!
schools_dat <- list(J = 8,
y = c(28, 8, -3, 7, -1, 1, 18, 12),
sigma = c(15, 10, 16, 11, 9, 11, 10, 18))

fit <- pstan(file = '8schools.stan', data = schools_dat, iter = 1000, chains = 4)

Which prints the long error message:

Error in compileCode(f, code, language = language, verbose = verbose) :
Compilation ERROR, function(s)/method(s) not created! In file included from C:/Users/MOESCHIA/Programs/R/R-3.2.2/library/rstan/include/rstan/stan_fit.hpp:80:0,
from C:/Users/MOESCHIA/Programs/R/R-3.2.2/library/rstan/include/rstan/rstaninc.hpp:3,
from file268863157f1d.cpp:444:
C:/Users/MOESCHIA/Programs/R/R-3.2.2/library/rstan/include/rstan/io/rlist_ref_var_context_factory.hpp:4:63: fatal error: stan/interface/var_context_factory/dump_factory.hpp: No such file or directory
compilation terminated.
make: *** [file268863157f1d.o] Error 1
Warning message:
the execution of the command 'make -f "C:/Users/MOESCHIA/Programs/R/R-32~1.2/etc/x64/Makeconf" -f "C:/Users/MOESCHIA/Programs/R/R-32~1.2/share/make/winshlib.mk" -f "//RKAMSCOE64/U_moeschia$/My Documents/.R/Makevars" SHLIB_LDFLAGS='$(SHLIB_CXXLDFLAGS)' SHLIB_LD='$(SHLIB_CXXLD)' SHLIB="file268863157f1d.dll" WIN=64 TCLBIN=64 OBJECTS="file268863157f1d.o"' sent a status 2
In addition: Warning message:
running command 'C:/Users/MOESCHIA/Programs/R/R-32~1.2/bin/x64/R CMD SHLIB file268863157f1d.cpp 2> file268863157f1d.cpp.err.txt' had status 1

I think there is a possibility that the problem comes from this part:

-f "//RKAMSCOE64/U_moeschia$/My Documents/.R/Makevars"

Because this is an external hard-drive and I don't use it. The computer belongs to the firm where I work and this might be related to some admin rights problem. I need to precise that whenever I tried to install something, I did it with right click, execute as administrator.

I don't know if I'm reporting my problem in the good section. If not, can you tell me where I should?

Thanks in advance for your help, I'm running out of solution after reinstalling R.



AND THEN:

I also tried the installation on my personal computer. I get the same final error:

Error in compileCode(f, code, language = language, verbose = verbose) :
Compilation ERROR, function(s)/method(s) not created! In file included from C:/Program Files/R/R-3.2.2/library/rstan/include/rstan/stan_fit.hpp:80:0,
from C:/Program Files/R/R-3.2.2/library/rstan/include/rstan/rstaninc.hpp:3,
from file1474331f1f.cpp:444:
C:/Program Files/R/R-3.2.2/library/rstan/include/rstan/io/rlist_ref_var_context_factory.hpp:4:63: fatal error: stan/interface/var_context_factory/dump_factory.hpp: No such file or directory
compilation terminated.
make: *** [file1474331f1f.o] Error 1
Warning message:
l'exécution de la commande 'make -f "C:/PROGRA~1/R/R-32~1.2/etc/x64/Makeconf" -f "C:/PROGRA~1/R/R-32~1.2/share/make/winshlib.mk" -f "D:/Users/P1XX3L-HP/Documents/.R/Makevars" SHLIB_LDFLAGS='$(SHLIB_CXXLDFLAGS)' SHLIB_LD='$(SHLIB_CXXLD)' SHLIB="file1474331f1f.dll" WIN=64 TCLBIN=64 OBJECTS="file1474331f1f.o"' renvoie un statut 2
In addition: Warning message:
running command 'C:/PROGRA~1/R/R-32~1.2/bin/x64/R CMD SHLIB file1474331f1f.cpp 2> file1474331f1f.cpp.err.txt' had status 1

> On Sep 18, 2015, at 2:48 AM, Jonathan Gilligan <notifi...@github.com> wrote:
>
> CRAN just updated StanHeaders to 2.8.0, but rstan is at 2.7.0-1 That is probably. at the heart of this.
>
> —
> Reply to this email directly or view it on GitHub.
>

[Jonah]

Yup, already responded on GitHub.

[Ben Goodrich]

On Friday, September 18, 2015 at 2:04:11 PM UTC-4, Jonah wrote:

Yup, already responded on GitHub.

Can someone with a Mac try (re)installing rstan from develop after installing StanHeaders, ggplot2, and gridExtra from CRAN? Then try manipulating the following things while doing one of the stan_demo() models

• specifying cores = 1 vs. cores > 1
• specifying open_progress = FALSE vs open_progress = TRUE
  
• inside RStudio vs. in Terminal
  

The only error among the 8 situations should be with cores > 1, open_progress = TRUE, and rstudio.

Thanks,
Ben

[Jonah Sol Gabry]

I can do this

--
You received this message because you are subscribed to a topic in the Google Groups "stan development mailing list" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/stan-dev/-UYHa9-K72c/unsubscribe.
To unsubscribe from this group and all its topics, send an email to stan-dev+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Jonah]

Hey Ben,

The only error is the one you anticipated. For all the other scenarios everything runs smoothly.

Jonah

On Friday, September 18, 2015 at 2:46:29 PM UTC-4, Jonah wrote:
> I can do this
>
>
> On Fri, Sep 18, 2015 at 2:43 PM, Ben Goodrich <goodri...@gmail.com> wrote:
>
> On Friday, September 18, 2015 at 2:04:11 PM UTC-4, Jonah wrote:Yup, already responded on GitHub.
>
> Can someone with a Mac try (re)installing rstan from develop after installing StanHeaders, ggplot2, and gridExtra from CRAN? Then try manipulating the following things while doing one of the stan_demo() models

> specifying cores = 1 vs. cores > 1specifying open_progress = FALSE vs open_progress = TRUE