Error when specyfing subjects in ctStanDiscretePars

[Kyabdro]

I'm trying to plot specific subjects DRIFT matrix, however when specifying any integer in the subjects argument, this error is produced:

Error in if ("all" %in% subjects || dim(e[[matname]])[2] == 1) { :
missing value where TRUE / FALSE is required

I've noticed that the corresponding conditional part of the script is made to change the names of the object e from "DRIFT" to "pop_DRIFT" and so on:

for (matname in c("DRIFT", "DIFFUSIONcov", "asymDIFFUSION")) {
    if ("all" %in% subjects || dim(e[[matname]])[2] == 1) {
      ctpars[[matname]] <- e[[paste0("pop_", matname)]]
    }

I think this error is due to an uptade of ctExtract which does not give names such as "DRIFT" anymore but now directly give "pop_DIRFT", etc... I think the error might be resolved easily by modfying the script accordingly:

for (matname in c("pop_DRIFT", "pop_DIFFUSIONcov", "pop_asymDIFFUSION"))

Then deleting the first conditional part that is not usefull anymore.

Did I guessed rigth?

[cdriveraus]

You're right, it's a problem caused by some other changes, I'll try fix it or post a solution soon. The pop matrices represent the mean though, not the subject pars.

[Kyabdro]

Did you made some changes ?

FYI - I've tried to install the most recent version from the github repo, however I have this issue (using both functions mentionned in the package description) :

installation of package ‘C:/Users/NPELLE~1/AppData/Local/Temp/RtmpiEtlOu/file4e802572663f/ctsem_3.3.12.tar.gz’ had non-zero exit status

I'm not shure if this is a problem from me or from the most recent version of the package...

[cdriveraus]

you caught me in the middle of a significant re-working sorry. hopefully another day and it's sorted. one of the recent commits had some errors, this should be fixed, but won't fix your issue.

[cdriveraus]

it still needs a little refining (certain cases are unnecessarily slow) but this feature is working properly again in the latest github release. You may need to re-fit any models, or you can try the ctStanFitUpdate() function. If fitting models with this intention, it is probably worth setting the argument savesubjectmatrices=TRUE when calling ctStanFit . This is off by default now because some users had memory trouble when running large models...

[Kyabdro]

I've still the same problem with the package installation from github. Maybe I caught you again in a significant re-working =)

[cdriveraus]

Can you post the model and fit code? This element is more complex these days because of differences between the full sampling procedure and optimization, but for my tests it is working so I'm clearly missing something.

[Kyabdro]

Sorry I was not clear : my problem is the installtion of the package from github :

installation of package ‘C:/Users/NPELLE~1/AppData/Local/Temp/RtmpiEtlOu/file4e802572663f/ctsem_3.3.12.tar.gz’ had non-zero exit status

So I did not have the chance to test the function again.

[cdriveraus]

I see. Other machines have been building from that commit, but I pushed another update through in any case. Could you try again from a fresh R session -- either using the install script:
source(file = 'https://github.com/cdriveraus/ctsem/raw/master/installctsem.R')

or if you've already run it recently, just:
remotes::install_github('cdriveraus/ctsem', INSTALL_opts = "--no-multiarch", dependencies = c("Depends", "Imports"))

if there are still errors, please post the complete install output, it should show a more meaningful error somewhere.

[Kyabdro]

Not working again :

Downloading GitHub repo cdriveraus/ctsem@HEAD

√ checking for file 'C:\Users\NPellerin\AppData\Local\Temp\Rtmp0Qvx52\remotes1c0c36dc3c24\cdriveraus-ctsem-85f3213/DESCRIPTION' (413ms)

• preparing 'ctsem': (2s)
  √ checking DESCRIPTION meta-information ...
• cleaning src
• checking for LF line-endings in source and make files and shell scripts (1.2s)
• checking for empty or unneeded directories
• building 'ctsem_3.4.0.tar.gz'

Installing package into ‘C:/Users/NPellerin/OneDrive/Documents/R/win-library/4.0’
(as ‘lib’ is unspecified)

• installing source package 'ctsem' ...
  ** using staged installation
  ** libs
  "C:/PROGRA1/R/R-401.2/bin/x64/Rscript" -e "source(file.path('..', 'tools', 'make_cc.R')); make_cc(commandArgs(TRUE))" stan_files/ctsmgen.stan
  "C:/PROGRA1/R/R-401.2/bin/x64/Rscript" -e "source(file.path('..', 'tools', 'make_cc.R')); make_cc(commandArgs(TRUE))" stan_files/cov.stan
  "C:/PROGRA1/R/R-401.2/bin/x64/Rscript" -e "source(file.path('..', 'tools', 'make_cc.R')); make_cc(commandArgs(TRUE))" stan_files/ctsm.stan

DIAGNOSTIC(S) FROM PARSER:
Info: integer division implicitly rounds to integer. Found int division: d * d - d / 2
Positive values rounded down, negative values rounded up or down in platform-dependent way.
Info: integer division implicitly rounds to integer. Found int division: d * d - d / 2
Positive values rounded down, negative values rounded up or down in platform-dependent way.

Wrote C++ file "stan_files/cov.cc"
"C:/rtools40/usr/bin/mingw64/bin/"g++ -m64 -std=c++1y -std=gnu++14 -I"C:/PROGRA1/R/R-401.2/include" -DNDEBUG -I"../inst/include" -I"C:/Users/NPellerin/OneDrive/Documents/R/win-library/4.0/StanHeaders/include/src" -DBOOST_DISABLE_ASSERTS -DEIGEN_NO_DEBUG -DRCPP_PARALLEL_USE_TBB=1 -I'C:/Users/NPellerin/OneDrive/Documents/R/win-library/4.0/BH/include' -I'C:/Users/NPellerin/OneDrive/Documents/R/win-library/4.0/Rcpp/include' -I'C:/Users/NPellerin/OneDrive/Documents/R/win-library/4.0/RcppEigen/include' -I'C:/Users/NPellerin/OneDrive/Documents/R/win-library/4.0/RcppParallel/include' -I'C:/Users/NPellerin/OneDrive/Documents/R/win-library/4.0/rstan/include' -I'C:/Users/NPellerin/OneDrive/Documents/R/win-library/4.0/StanHeaders/include' -DRCPP_PARALLEL_USE_TBB=1 -I"C:/Users/NPellerin/OneDrive/Documents/R/win-library/4.0/RcppParallel/include" -D_REENTRANT -DSTAN_THREADS -mtune=native -O1 -Wno-ignored-attributes -Wno-deprecated-declarations -Wno-ignored-attributes -Wno-deprecated-declarations -c init.cpp -o init.o
sh: C:/rtools40/usr/bin/mingw64/bin/g++: No such file or directory
make: *** [C:/PROGRA1/R/R-401.2/etc/x64/Makeconf:229: init.o] Error 127
make: *** Waiting for unfinished jobs....
Wrote C++ file "stan_files/ctsmgen.cc"
Wrote C++ file "stan_files/ctsm.cc"
rm stan_files/ctsmgen.cc stan_files/ctsm.cc stan_files/cov.cc
ERROR: compilation failed for package 'ctsem'

• removing 'C:/Users/NPellerin/OneDrive/Documents/R/win-library/4.0/ctsem'
• restoring previous 'C:/Users/NPellerin/OneDrive/Documents/R/win-library/4.0/ctsem'
  Warning message:
  In i.p(...) :
  installation of package ‘C:/Users/NPELLE~1/AppData/Local/Temp/Rtmp0Qvx52/file1c0ca8c7f6a/ctsem_3.4.0.tar.gz’ had non-zero exit status

[cdriveraus]

Ok. You're on windows it looks like? rstan is not always so easy to get compiling, for some reason. There is some discussion about what I 'think' is the issue here:
https://discourse.mc-stan.org/t/error-in-file-con-r-cannot-open-the-connection-c-rtools40-usr-mingw-bin-g-not-found/17310/2

The first thing to check seems to be the .Renviron file, which, if it exists, should be in your home directory. Make sure it doesn't have a BINPREF line, if it does, remove it.
https://stackoverflow.com/questions/40788645/how-to-create-renviron-file

[Kyabdro]

I don't have any BINPREF line, the only line present in .Renviron is

PATH="${RTOOLS40_HOME}\usr\bin;${PATH}"

Note that I had no trouble to install previous versions of ctsem, including the actual CRAN version. I'm not shure if this is an issue related to Rstan.

That said, I will explore the link you gave, and do what they propose to see if it resolve the problem.

[Kyabdro]

Ok, I reinstalled Rtools, and now everything is ok. I'll try the function of interest of this issue and I come back to you.

[Kyabdro]

Now, if I specify one subject with ctStanDiscretePars (e.g., subjects = 4) this work fine.

However specifying multiple subject (e.g., subjects = c(2,4)) does not work. I've tried to use as.integer to the vector of subjects but the same error occurs anyway :

Error : $ operator is invalid for atomic vectors

[cdriveraus]

Sorry! that should be fixed too now. Classic R sloppy typing issue. Here's the test script I was using, just in case of curiosity :)

dri <- sort(-log1p(exp(rnorm(10,0,2))))

for(i in 1:10){
gm <- ctModel(LAMBDA=diag(1), DRIFT = dri[i],DIFFUSION = .1,MANIFESTVAR = .2,Tpoints=30,T0MEANS = 30)
d <- data.frame(ctGenerate(ctmodelobj = gm,n.subjects = 1,dtmean = .1),dri=dri[i])
d$id <- i
if(i==1) dat <- d else dat <- rbind(dat,d)
}
plot(dat$time,dat$Y1)

m <- ctModel(LAMBDA=diag(1),type='stanct',
  DRIFT = 'dr||||dri',
  TIpredNames = 'dri',tipredDefault = F)

f=ctStanFit(datalong = dat,ctstanmodel = m,plot=F,cores=1,
  savesubjectmatrices = T,verbose=1)



e=ctExtract(f,subjectMatrices = T,cores=6)
plot(e$subj_DRIFT[1,,,],dry)
e$subj_DRIFT[1,,,]
ctStanDiscretePars(ctstanfitobj = f,plot=T,subjects = 1)
ctStanDiscretePars(ctstanfitobj = f,plot=T,subjects = 10)
ctStanDiscretePars(ctstanfitobj = f,plot=T,subjects = c(1,10))
ctKalman(f,plot=T,subjects = 1:10,removeObs = T)

[Kyabdro]

Everything works fine now, thank you very much =)