Compilation error with brms in Ubuntu Minconda3 environment

sree_d · October 24, 2020, 10:57pm

OS: Ubuntu 20.04 64bit

Environment: Miniconda3 with R-4.0.3 in a conda environment using Conda built-in compilers

I’ve set up brms and rstan in an environment under conda to work with R-4.0.3

I could run the first iteration of the model (see the code below) using the default adapt_delta at 0.8. It gave me warnings about divergences (I expected this but wanted a baselines). I modified the adapt_delta to 0.95 and when I attempted to rerun the model the compilation stopped and I got the following error, which i cannot understand.

If anyone here can shed some light on the problem source that would be great. Here is the code and outut in its’ entirety for the first and second runs.

Why did the first model succeed in its’ compilation but the second one failed?

First run (Compilation Successful)

library(brms)
Loading required package: Rcpp
Loading ‘brms’ package (version 2.13.5). Useful instructions
can be found by typing help(‘brms’). A more detailed introduction
to the package is available through vignette(‘brms_overview’).

Attaching package: ‘brms’

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

ar

mod1 ← brm(ln_fatalities ~ ln_beer+ln_population+ln_totalvmt +

          (1|state) + (1|primary) + (1|year), data=trafsaf,

         cores=2, chains=2, control = list(adapt_delta = 0.80))

Compiling Stan program…
Start sampling

SAMPLING FOR MODEL ‘ded8ea9a2b8d5f29ec0f3be9157f5983’ NOW (CHAIN 1).
Chain 1:
Chain 1: Gradient evaluation took 0.000568 seconds
Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 5.68 seconds.
Chain 1: Adjust your expectations accordingly!
Chain 1:
Chain 1:

SAMPLING FOR MODEL ‘ded8ea9a2b8d5f29ec0f3be9157f5983’ NOW (CHAIN 2).
Chain 1: Iteration: 1 / 2000 [ 0%] (Warmup)
Chain 2:
Chain 2: Gradient evaluation took 0.000543 seconds
Chain 2: 1000 transitions using 10 leapfrog steps per transition would take 5.43 seconds.
Chain 2: Adjust your expectations accordingly!
Chain 2:
Chain 2:
Chain 2: Iteration: 1 / 2000 [ 0%] (Warmup)
Chain 1: Iteration: 200 / 2000 [ 10%] (Warmup)
Chain 2: Iteration: 200 / 2000 [ 10%] (Warmup)
Chain 1: Iteration: 400 / 2000 [ 20%] (Warmup)
Chain 1: Iteration: 600 / 2000 [ 30%] (Warmup)
Chain 2: Iteration: 400 / 2000 [ 20%] (Warmup)
Chain 1: Iteration: 800 / 2000 [ 40%] (Warmup)
Chain 1: Iteration: 1000 / 2000 [ 50%] (Warmup)
Chain 1: Iteration: 1001 / 2000 [ 50%] (Sampling)
Chain 2: Iteration: 600 / 2000 [ 30%] (Warmup)
Chain 1: Iteration: 1200 / 2000 [ 60%] (Sampling)
Chain 1: Iteration: 1400 / 2000 [ 70%] (Sampling)
Chain 2: Iteration: 800 / 2000 [ 40%] (Warmup)
Chain 1: Iteration: 1600 / 2000 [ 80%] (Sampling)
Chain 2: Iteration: 1000 / 2000 [ 50%] (Warmup)
Chain 2: Iteration: 1001 / 2000 [ 50%] (Sampling)
Chain 1: Iteration: 1800 / 2000 [ 90%] (Sampling)
Chain 2: Iteration: 1200 / 2000 [ 60%] (Sampling)
Chain 1: Iteration: 2000 / 2000 [100%] (Sampling)
Chain 1:
Chain 1: Elapsed Time: 93.8875 seconds (Warm-up)
Chain 1: 77.7345 seconds (Sampling)
Chain 1: 171.622 seconds (Total)
Chain 1:
Chain 2: Iteration: 1400 / 2000 [ 70%] (Sampling)
Chain 2: Iteration: 1600 / 2000 [ 80%] (Sampling)
Chain 2: Iteration: 1800 / 2000 [ 90%] (Sampling)
Chain 2: Iteration: 2000 / 2000 [100%] (Sampling)
Chain 2:
Chain 2: Elapsed Time: 151.723 seconds (Warm-up)
Chain 2: 82.9083 seconds (Sampling)
Chain 2: 234.631 seconds (Total)
Chain 2:
Warning messages:
1: There were 242 divergent transitions after warmup. See
http://mc-stan.org/misc/warnings.html#divergent-transitions-after-warmup
to find out why this is a problem and how to eliminate them.
2: There were 161 transitions after warmup that exceeded the maximum treedepth. Increase max_treedepth above 10. Se
e
http://mc-stan.org/misc/warnings.html#maximum-treedepth-exceeded
3: Examine the pairs() plot to diagnose sampling problems

4: Bulk Effective Samples Size (ESS) is too low, indicating posterior means and medians may be unreliable.
Running the chains for more iterations may help. See
http://mc-stan.org/misc/warnings.html#bulk-ess
5: Tail Effective Samples Size (ESS) is too low, indicating posterior variances and tail quantiles may be unreliabl
e.
Running the chains for more iterations may help. See
http://mc-stan.org/misc/warnings.html#tail-ess

Second Run (Compilation Fails,ONly Change is adapt_delta)

mod1 ← brm(ln_fatalities ~ ln_beer+ln_population+ln_totalvmt +

          (1|state) + (1|primary) + (1|year), data=trafsaf,

         cores=2, chains=2, control = list(adapt_delta = 0.95))

Compiling Stan program…
make cmd is
make -f ‘/home/sreezach/miniconda3/envs/asd/lib/R/etc/Makeconf’ -f ‘/home/sreezach/miniconda3/envs/asd/lib/R/shar
e/make/shlib.mk’ CXX=‘(CXX14) (CXX14STD)’ CXXFLAGS=‘(CXX14FLAGS)' CXXPICFLAGS='(CXX14PICFLAGS)’ SHLIB_LDFLAGS=’
(SHLIB_CXX14LDFLAGS)' SHLIB_LD='(SHLIB_CXX14LD)’ SHLIB=‘file17c1267cfeb5.so’ OBJECTS=‘file17c1267cfeb5.o’

make would use
x86_64-conda-linux-gnu-c++ -std=gnu++14 -I"/home/sreezach/miniconda3/envs/asd/lib/R/include" -DNDEBUG -I"/home/sr
eezach/miniconda3/envs/asd/lib/R/library/Rcpp/include/" -I"/home/sreezach/miniconda3/envs/asd/lib/R/library/RcppEi
gen/include/" -I"/home/sreezach/miniconda3/envs/asd/lib/R/library/RcppEigen/include/unsupported" -I"/home/sreezac
h/miniconda3/envs/asd/lib/R/library/BH/include" -I"/home/sreezach/miniconda3/envs/asd/lib/R/library/StanHeaders/inc
lude/src/" -I"/home/sreezach/miniconda3/envs/asd/lib/R/library/StanHeaders/include/" -I"/home/sreezach/miniconda3
/envs/asd/lib/R/library/RcppParallel/include/" -I"/home/sreezach/miniconda3/envs/asd/lib/R/library/rstan/include"
-DEIGEN_NO_DEBUG -DBOOST_DISABLE_ASSERTS -DBOOST_PENDING_INTEGER_LOG2_HPP -DSTAN_THREADS -DBOOST_NO_AUTO_PTR -
include ‘/home/sreezach/miniconda3/envs/asd/lib/R/library/StanHeaders/include/stan/math/prim/mat/fun/Eigen.hpp’ -D
_REENTRANT -DRCPP_PARALLEL_USE_TBB=1 -DNDEBUG -D_FORTIFY_SOURCE=2 -O2 -isystem /home/sreezach/miniconda3/envs/asd
/include -I/home/sreezach/miniconda3/envs/asd/include -Wl,-rpath-link,/home/sreezach/miniconda3/envs/asd/lib -fpi
c -fvisibility-inlines-hidden -fmessage-length=0 -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-prot
ector-strong -fno-plt -O2 -ffunction-sections -pipe -isystem /home/sreezach/miniconda3/envs/asd/include -fdebug-pre
fix-map=/home/conda/feedstock_root/build_artifacts/r-base_1603051729662/work=/usr/local/src/conda/r-base-4.0.3 -fde
bug-prefix-map=/home/sreezach/miniconda3/envs/asd=/usr/local/src/conda-prefix -c file17c1267cfeb5.cpp -o file17c12
67cfeb5.o
if test “zfile17c1267cfeb5.o” != “z”; then
echo x86_64-conda-linux-gnu-c++ -std=gnu++14 -shared -L"/home/sreezach/miniconda3/envs/asd/lib/R/lib" -Wl,-O2 -Wl
,–sort-common -Wl,–as-needed -Wl,-z,relro -Wl,-z,now -Wl,–disable-new-dtags -Wl,–gc-sections -Wl,-rpath,/home/s
reezach/miniconda3/envs/asd/lib -Wl,-rpath-link,/home/sreezach/miniconda3/envs/asd/lib -L/home/sreezach/miniconda3/
envs/asd/lib -Wl,-rpath-link,/home/sreezach/miniconda3/envs/asd/lib -o file17c1267cfeb5.so file17c1267cfeb5.o ‘/ho
me/sreezach/miniconda3/envs/asd/lib/R/library/rstan/lib//libStanServices.a’ -L’/home/sreezach/miniconda3/envs/asd/l
ib/R/library/StanHeaders/lib/’ -lStanHeaders -L’/home/sreezach/miniconda3/envs/asd/lib/R/library/RcppParallel/lib/’
-ltbb -L"/home/sreezach/miniconda3/envs/asd/lib/R/lib" -lR;
x86_64-conda-linux-gnu-c++ -std=gnu++14 -shared -L"/home/sreezach/miniconda3/envs/asd/lib/R/lib" -Wl,-O2 -Wl,–so
rt-common -Wl,–as-needed -Wl,-z,relro -Wl,-z,now -Wl,–disable-new-dtags -Wl,–gc-sections -Wl,-rpath,/home/sreeza
ch/miniconda3/envs/asd/lib -Wl,-rpath-link,/home/sreezach/miniconda3/envs/asd/lib -L/home/sreezach/miniconda3/envs/
asd/lib -Wl,-rpath-link,/home/sreezach/miniconda3/envs/asd/lib -o file17c1267cfeb5.so file17c1267cfeb5.o ‘/home/sr
eezach/miniconda3/envs/asd/lib/R/library/rstan/lib//libStanServices.a’ -L’/home/sreezach/miniconda3/envs/asd/lib/R/
library/StanHeaders/lib/’ -lStanHeaders -L’/home/sreezach/miniconda3/envs/asd/lib/R/library/RcppParallel/lib/’ -ltb
b -L"/home/sreezach/miniconda3/envs/asd/lib/R/lib" -lR;
fi

sree_d · October 24, 2020, 11:25pm

I wanted to add additional output I’m seeing running brms model (same as before)

Run 3 gives an error in one of the chains, suggests I consider running a single chain)

mod1 ← brm(ln_fatalities ~ ln_beer+ln_population+ln_totalvmt +

          (1|state) + (1|primary) + (1|year), data=trafsaf,

         cores=2, chains=2, control=(adapt_delta=0.95))

Compiling Stan program…
Start sampling
Error in config_argss(chains = chains, iter = iter, warmup = warmup, thin = thin, :
‘control’ should be a named list
Error in config_argss(chains = chains, iter = iter, warmup = warmup, thin = thin, :
‘control’ should be a named list
error in specifying arguments; sampling not done
error in specifying arguments; sampling not done
here are whatever error messages were returned
[[1]]
Stan model ‘ded8ea9a2b8d5f29ec0f3be9157f5983’ does not contain samples.

[[2]]
Stan model ‘ded8ea9a2b8d5f29ec0f3be9157f5983’ does not contain samples.

Warning message:
In .local(object, …) :
some chains had errors; consider specifying chains = 1 to debug

**Run 4 with chains=1 setting) stops at Selection: Selection: Output

mod1 ← brm(ln_fatalities ~ ln_beer+ln_population+ln_totalvmt +

          (1|state) + (1|primary) + (1|year), data=trafsaf,

         cores=2, chains=1, control=(adapt_delta=0.95))

Compiling Stan program…
Selection: Selection:

ahartikainen · October 28, 2020, 7:05am

Hi,

So

First run was succesful.
Second run failed (but no error message?)
Third run fails because you don’t use named list?

FYI:
Also, you can use three ` to get a code block

```
code here
```

then looks like

code here

sree_d · October 28, 2020, 1:27pm

Thanks for getting back @ahartikainen. The error I posted on the first message was displaying the errors I got in second run; Third one, as you said is because I did not use a named list. I rectified that. The original error(s) in my first message are resolved as of this morning. My compilation was failing and behaving in an haphazard due to one of memory sticks failing and causing compiler and seg-fault errors. I replaced with a new memory stick and my brms / Stan code is working again.

Thanks for the tip on how to post code block.

Sree

Topic		Replies	Views
Compiling errors after previous successful runs brms	3	589	July 18, 2022
Sampler won't start if adapt_delta is used: why? Modeling rstan , fitting-issues , brms	6	1164	September 16, 2021
Error in Stan Code brms rstan , performance	7	1125	May 8, 2020
C Stack Trace Errors brms rstan	6	1500	November 9, 2020
Cannot run models after updating brms and rstan brms	31	4993	October 2, 2020

Compilation error with brms in Ubuntu Minconda3 environment

Related topics