# Brms hangs after drawing samples

**URL:** <https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262>\
**Category:** Interfaces\
**Tags:** brms\
**Created:** [February 8, 2022, 3:29am UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262 "2022-02-08T03:29:51Z")\
**Posts on this page:** 11\
**Page:** 1

<div class="post-metadata">

**Author:** ![matti](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/matti/32/13631_2.png) [@matti](https://discourse.mc-stan.org/u/matti)\
**Post date:** [February 8, 2022, 3:29am UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262/1 "2022-02-08T03:29:51Z")

</div>

Hi!

This is related to [my other question](https://discourse.mc-stan.org/t/understanding-brms-save-pars/26261) about `save_pars()` and large multinomial models. I am trying to draw samples from a multinomial model with correlated random intercepts across categories. I can do this with smaller numbers of groups (under 1000), but with larger numbers of groups, the model samples but then hangs and never returns the fitted model object.

EDIT: I can kind of fix this by not saving the “random effects” as discussed in [my other question](https://discourse.mc-stan.org/t/understanding-brms-save-pars/26261). But there are cases when I would like to a) use the cmdstanr backend, and b) look at the group-level effects. So I’m leaving this here in case anyone has thoughts about it.

It’s a bit hard to give a reproducible example of the issue, but here is code that works.

```r
library(brms)
#> Loading required package: Rcpp
#> Loading 'brms' package (version 2.16.7). 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
library(tictoc)

# Number of groups
N <- 500
dat <- data.frame(
  id = factor(1:N),
  y1 = rbinom(N, 100, 0.1), 
  y2 = rbinom(N, 100, 0.2), 
  y3 = rbinom(N, 100, 0.3),
  y4 = rbinom(N, 100, 0.4), 
  y5 = rbinom(N, 100, 0.5), 
  y6 = rbinom(N, 100, 0.6)
)
dat$size <- with(dat, y1 + y2 + y3 + y4 + y5 + y6)

tic()
fit <- brm(
  bf(cbind(y1, y2, y3, y4, y5, y6) | trials(size) ~ 1 + (1 |p| id)), 
  family = multinomial(),
  data = dat, 
  save_pars = save_pars(group = FALSE),
  iter = 10, chains = 1,
  backend = "cmdstanr"
)
#> Start sampling
#> Running MCMC with 1 chain...
#> 
#> Chain 1 WARNING: No variance estimation is 
#> Chain 1 performed for num_warmup < 20 
#> Chain 1 Iteration: 1 / 10 [10%] (Warmup) 
#> Chain 1 Iteration: 6 / 10 [60%] (Sampling) 
#> Chain 1 Iteration: 10 / 10 [100%] (Sampling) 
#> Chain 1 finished in 0.0 seconds.
#> 
#> Warning: 5 of 5 (100.0%) transitions ended with a divergence.
#> This may indicate insufficient exploration of the posterior distribution.
#> Possible remedies include: 
#> * Increasing adapt_delta closer to 1 (default is 0.8) 
#> * Reparameterizing the model (e.g. using a non-centered parameterization)
#> * Using informative or weakly informative prior distributions
toc()
#> 13.153 sec elapsed

```

Created on 2022-02-08 by the [reprex package](https://reprex.tidyverse.org) (v2.0.1)

> **Session info**
>
> ```r
> sessioninfo::session_info()
> #> ─ Session info ───────────────────────────────────────────────────────────────
> #> setting value
> #> version R version 4.1.2 (2021-11-01)
> #> os macOS Monterey 12.1
> #> system aarch64, darwin20
> #> ui X11
> #> language (EN)
> #> collate en_US.UTF-8
> #> ctype en_US.UTF-8
> #> tz Asia/Seoul
> #> date 2022-02-08
> #> pandoc 2.17.1.1 @ /Applications/RStudio.app/Contents/MacOS/quarto/bin/ (via rmarkdown)
> #> 
> #> ─ Packages ───────────────────────────────────────────────────────────────────
> #> package * version date (UTC) lib source
> #> abind 1.4-5 2016-07-21 [1] CRAN (R 4.1.0)
> #> assertthat 0.2.1 2019-03-21 [1] CRAN (R 4.1.0)
> #> backports 1.4.1 2021-12-13 [1] CRAN (R 4.1.1)
> #> base64enc 0.1-3 2015-07-28 [1] CRAN (R 4.1.0)
> #> bayesplot 1.8.1 2021-06-14 [1] CRAN (R 4.1.0)
> #> bridgesampling 1.1-2 2021-04-16 [1] CRAN (R 4.1.0)
> #> brms * 2.16.7 2022-02-08 [1] Github (paul-buerkner/brms@3164328)
> #> Brobdingnag 1.2-7 2022-02-03 [1] CRAN (R 4.1.1)
> #> callr 3.7.0 2021-04-20 [1] CRAN (R 4.1.0)
> #> checkmate 2.0.0 2020-02-06 [1] CRAN (R 4.1.1)
> #> cli 3.1.1 2022-01-20 [1] CRAN (R 4.1.2)
> #> cmdstanr 0.4.0.9001 2022-01-25 [1] Github (stan-dev/cmdstanr@a2a97d9)
> #> coda 0.19-4 2020-09-30 [1] CRAN (R 4.1.0)
> #> codetools 0.2-18 2020-11-04 [1] CRAN (R 4.1.2)
> #> colorspace 2.0-2 2021-06-24 [1] CRAN (R 4.1.1)
> #> colourpicker 1.1.1 2021-10-04 [1] CRAN (R 4.1.1)
> #> crayon 1.4.2 2021-10-29 [1] CRAN (R 4.1.1)
> #> crosstalk 1.2.0 2021-11-04 [1] CRAN (R 4.1.1)
> #> curl 4.3.2 2021-06-23 [1] CRAN (R 4.1.0)
> #> data.table 1.14.2 2021-09-27 [1] CRAN (R 4.1.1)
> #> DBI 1.1.2 2021-12-20 [1] CRAN (R 4.1.1)
> #> digest 0.6.29 2021-12-01 [1] CRAN (R 4.1.1)
> #> distributional 0.3.0 2022-01-05 [1] CRAN (R 4.1.1)
> #> dplyr 1.0.7 2021-06-18 [1] CRAN (R 4.1.0)
> #> DT 0.20 2021-11-15 [1] CRAN (R 4.1.1)
> #> dygraphs 1.1.1.6 2018-07-11 [1] CRAN (R 4.1.0)
> #> ellipsis 0.3.2 2021-04-29 [1] CRAN (R 4.1.0)
> #> emmeans 1.7.2 2022-01-04 [1] CRAN (R 4.1.1)
> #> estimability 1.3 2018-02-11 [1] CRAN (R 4.1.0)
> #> evaluate 0.14 2019-05-28 [1] CRAN (R 4.1.0)
> #> fansi 1.0.2 2022-01-14 [1] CRAN (R 4.1.1)
> #> farver 2.1.0 2021-02-28 [1] CRAN (R 4.1.0)
> #> fastmap 1.1.0 2021-01-25 [1] CRAN (R 4.1.0)
> #> fs 1.5.2 2021-12-08 [1] CRAN (R 4.1.1)
> #> generics 0.1.2 2022-01-31 [1] CRAN (R 4.1.1)
> #> ggplot2 3.3.5 2021-06-25 [1] CRAN (R 4.1.1)
> #> ggridges 0.5.3 2021-01-08 [1] CRAN (R 4.1.1)
> #> glue 1.6.1 2022-01-22 [1] CRAN (R 4.1.2)
> #> gridExtra 2.3 2017-09-09 [1] CRAN (R 4.1.1)
> #> gtable 0.3.0 2019-03-25 [1] CRAN (R 4.1.1)
> #> gtools 3.9.2 2021-06-06 [1] CRAN (R 4.1.0)
> #> highr 0.9 2021-04-16 [1] CRAN (R 4.1.0)
> #> htmltools 0.5.2 2021-08-25 [1] CRAN (R 4.1.1)
> #> htmlwidgets 1.5.4 2021-09-08 [1] CRAN (R 4.1.1)
> #> httpuv 1.6.5 2022-01-05 [1] CRAN (R 4.1.1)
> #> igraph 1.2.11 2022-01-04 [1] CRAN (R 4.1.1)
> #> inline 0.3.19 2021-05-31 [1] CRAN (R 4.1.0)
> #> jsonlite 1.7.3 2022-01-17 [1] CRAN (R 4.1.2)
> #> knitr 1.37 2021-12-16 [1] CRAN (R 4.1.1)
> #> later 1.3.0 2021-08-18 [1] CRAN (R 4.1.1)
> #> lattice 0.20-45 2021-09-22 [1] CRAN (R 4.1.2)
> #> lifecycle 1.0.1 2021-09-24 [1] CRAN (R 4.1.1)
> #> loo 2.4.1 2020-12-09 [1] CRAN (R 4.1.0)
> #> magrittr 2.0.2 2022-01-26 [1] CRAN (R 4.1.1)
> #> markdown 1.1 2019-08-07 [1] CRAN (R 4.1.0)
> #> Matrix 1.4-0 2021-12-08 [1] CRAN (R 4.1.1)
> #> matrixStats 0.61.0 2021-09-17 [1] CRAN (R 4.1.1)
> #> mime 0.12 2021-09-28 [1] CRAN (R 4.1.1)
> #> miniUI 0.1.1.1 2018-05-18 [1] CRAN (R 4.1.0)
> #> munsell 0.5.0 2018-06-12 [1] CRAN (R 4.1.0)
> #> mvtnorm 1.1-3 2021-10-08 [1] CRAN (R 4.1.1)
> #> nlme 3.1-155 2022-01-13 [1] CRAN (R 4.1.1)
> #> pillar 1.7.0 2022-02-01 [1] CRAN (R 4.1.1)
> #> pkgbuild 1.3.1 2021-12-20 [1] CRAN (R 4.1.1)
> #> pkgconfig 2.0.3 2019-09-22 [1] CRAN (R 4.1.0)
> #> plyr 1.8.6 2020-03-03 [1] CRAN (R 4.1.0)
> #> posterior 1.2.0 2022-01-05 [1] CRAN (R 4.1.1)
> #> prettyunits 1.1.1 2020-01-24 [1] CRAN (R 4.1.0)
> #> processx 3.5.2 2021-04-30 [1] CRAN (R 4.1.0)
> #> promises 1.2.0.1 2021-02-11 [1] CRAN (R 4.1.0)
> #> ps 1.6.0 2021-02-28 [1] CRAN (R 4.1.0)
> #> purrr 0.3.4 2020-04-17 [1] CRAN (R 4.1.0)
> #> R6 2.5.1 2021-08-19 [1] CRAN (R 4.1.1)
> #> Rcpp * 1.0.8 2022-01-13 [1] CRAN (R 4.1.1)
> #> RcppParallel 5.1.5 2022-01-05 [1] CRAN (R 4.1.1)
> #> reprex 2.0.1 2021-08-05 [1] CRAN (R 4.1.1)
> #> reshape2 1.4.4 2020-04-09 [1] CRAN (R 4.1.0)
> #> rlang 1.0.1 2022-02-03 [1] CRAN (R 4.1.1)
> #> rmarkdown 2.11 2021-09-14 [1] CRAN (R 4.1.1)
> #> rsconnect 0.8.25 2021-11-19 [1] CRAN (R 4.1.1)
> #> rstan 2.26.6 2022-01-25 [1] local
> #> rstantools 2.1.1 2020-07-06 [1] CRAN (R 4.1.0)
> #> rstudioapi 0.13 2020-11-12 [1] CRAN (R 4.1.0)
> #> scales 1.1.1 2020-05-11 [1] CRAN (R 4.1.0)
> #> sessioninfo 1.2.2 2021-12-06 [1] CRAN (R 4.1.1)
> #> shiny 1.7.1 2021-10-02 [1] CRAN (R 4.1.1)
> #> shinyjs 2.1.0 2021-12-23 [1] CRAN (R 4.1.1)
> #> shinystan 2.5.0 2018-05-01 [1] CRAN (R 4.1.0)
> #> shinythemes 1.2.0 2021-01-25 [1] CRAN (R 4.1.0)
> #> StanHeaders 2.26.6 2022-01-25 [1] local
> #> stringi 1.7.6 2021-11-29 [1] CRAN (R 4.1.1)
> #> stringr 1.4.0 2019-02-10 [1] CRAN (R 4.1.1)
> #> tensorA 0.36.2 2020-11-19 [1] CRAN (R 4.1.0)
> #> threejs 0.3.3 2020-01-21 [1] CRAN (R 4.1.0)
> #> tibble 3.1.6 2021-11-07 [1] CRAN (R 4.1.1)
> #> tictoc * 1.0.1 2021-04-19 [1] CRAN (R 4.1.0)
> #> tidyselect 1.1.1 2021-04-30 [1] CRAN (R 4.1.0)
> #> utf8 1.2.2 2021-07-24 [1] CRAN (R 4.1.0)
> #> V8 4.1.0 2022-02-06 [1] CRAN (R 4.1.2)
> #> vctrs 0.3.8 2021-04-29 [1] CRAN (R 4.1.0)
> #> withr 2.4.3 2021-11-30 [1] CRAN (R 4.1.1)
> #> xfun 0.29 2021-12-14 [1] CRAN (R 4.1.1)
> #> xtable 1.8-4 2019-04-21 [1] CRAN (R 4.1.0)
> #> xts 0.12.1 2020-09-09 [1] CRAN (R 4.1.0)
> #> yaml 2.2.2 2022-01-25 [1] CRAN (R 4.1.1)
> #> zoo 1.8-9 2021-03-09 [1] CRAN (R 4.1.0)
> #> 
> #> [1] /Library/Frameworks/R.framework/Versions/4.1-arm64/Resources/library
> #> 
> #> ──────────────────────────────────────────────────────────────────────────────
> 
> ```

However, running the same code as above but increasing N to, say, 5000 results in brms hanging after drawing the samples:

```
Compiling Stan program...
Start sampling
Running MCMC with 1 chain...

Chain 1 WARNING: No variance estimation is 
Chain 1 performed for num_warmup < 20 
Chain 1 Iteration: 1 / 10 [10%] (Warmup) 
Chain 1 Iteration: 6 / 10 [60%] (Sampling) 
Chain 1 Iteration: 10 / 10 [100%] (Sampling) 
Chain 1 finished in 0.3 seconds.

Warning: 5 of 5 (100.0%) transitions ended with a divergence.
This may indicate insufficient exploration of the posterior distribution.
Possible remedies include: 
  * Increasing adapt_delta closer to 1 (default is 0.8) 
  * Reparameterizing the model (e.g. using a non-centered parameterization)
  * Using informative or weakly informative prior distributions 

```

I’ve tried this with rstan and cmdstanr backends, but it just sits there. I left it in the background and even after an hour it wasn’t done.

I don’t think this is a memory issue, but would appreciate any pointers! Thanks.

---

<div class="post-metadata">

**Author:** ![Chris\_Statwonk\_Peters](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/chris_statwonk_peters/32/15237_2.png) [@Chris\_Statwonk\_Peters](https://discourse.mc-stan.org/u/Chris_Statwonk_Peters)\
**Post date:** [October 24, 2022, 12:46pm UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262/2 "2022-10-24T12:46:41Z")

</div>

I’ve run into this issue, too, but I don’t have a reproducible example just yet. In my case stan or `brms` hangs after drawing samples. Samples for all chains appear finished but the repl is busy and won’t return.

---

<div class="post-metadata">

**Author:** ![simon\_mills](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/simon_mills/32/17115_2.png) [@simon\_mills](https://discourse.mc-stan.org/u/simon_mills)\
**Post date:** [October 26, 2022, 7:02am UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262/3 "2022-10-26T07:02:01Z")

</div>

Which brms version are you using? After sampling has finished, brms reads the stan csv files back into R. Previously it did this via `rstan::read_stan_csv()`, which can be extremely slow once csv files get large. The current version (`2.18.0`) uses `cmdstanr::read_cmdstan_csv()` on the backend and doesn’t hang with large file sizes. I’m fairly confident this will have been the issue in the original post, which used `2.16.7`.

If you have a model that has completed sampling but is now stuck in this end step (or timed out etc.), then- assuming you know where they got saved- you can also rebuild the `brmsfit` object from the stan csv files.

---

<div class="post-metadata">

**Author:** ![t-pollington](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/t-pollington/32/14497_2.png) [@t-pollington](https://discourse.mc-stan.org/u/t-pollington)\
**Post date:** [November 30, 2022, 3:32pm UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262/4 "2022-11-30T15:32:27Z")

</div>

I’m getting a hang too but on v`2.18.0`. After a (successful) run after restarting R, the .rds filesize was 66MB. I’m using `save_pars = save_pars(all = TRUE)` so that I can later use the fitted object to calculate LOO, so perhaps the reason.

@Chris_Statwonk_Peters would a `gc()` command before model fitting help to avoid this? I had been working 10s of hours in the same R session, perhaps that’s why this intermittently happened to me?

Would be useful to add the ‘bugs’ tag to this @matti, as the thread author for better visibility to package maintainer.

OS: 5.10.16.3-microsoft-standard-WSL2

---

<div class="post-metadata">

**Author:** ![t-pollington](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/t-pollington/32/14497_2.png) [@t-pollington](https://discourse.mc-stan.org/u/t-pollington)\
**Post date:** [December 1, 2022, 5:21pm UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262/5 "2022-12-01T17:21:12Z")

</div>

Using `gc()` hasn’t helped unfortunately as it is hanging again.

@paul.buerkner: feature idea. Just like a progress readout is printed for the % progress of the chains, is it possible to track the progress of saving the fitted object & its parameters to `.rds`? The tricky bit for the user is determining if it’s crashed or just taking a long time to complete.

ps: `.rds` file expected to be ~ 106MB.

---

<div class="post-metadata">

**Author:** ![paul.buerkner](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/paul.buerkner/32/3303_2.png) [@paul.buerkner](https://discourse.mc-stan.org/u/paul.buerkner)\
**Post date:** [December 1, 2022, 5:30pm UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262/6 "2022-12-01T17:30:05Z")

</div>

this would have to be an rstan oder cmdstanr feature as brms does not control things that happen during the actual model fitting.

---

<div class="post-metadata">

**Author:** ![t-pollington](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/t-pollington/32/14497_2.png) [@t-pollington](https://discourse.mc-stan.org/u/t-pollington)\
**Post date:** [December 1, 2022, 9:46pm UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262/7 "2022-12-01T21:46:50Z")

</div>

Thanks have suggested it here [Feature suggestion: Printout progress when saving model fit object · Issue #725 · stan-dev/cmdstanr · GitHub](https://github.com/stan-dev/cmdstanr/issues/725).

---

<div class="post-metadata">

**Author:** ![joshuakrobertson](https://avatars.discourse-cdn.com/v4/letter/j/85e7bf/32.png) [@joshuakrobertson](https://discourse.mc-stan.org/u/joshuakrobertson)\
**Post date:** [May 22, 2023, 3:19pm UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262/8 "2023-05-22T15:19:37Z")

</div>

Hi everyone,

Facing the same problem here. Any new solutions? @Matti, how long have you left your model hanging for post sampling (I know you mentioned leaving it for an hour in your last post, but curious if you’ve tried longer).

I don’t see any movement following your GitHub suggestion @t-pollington…

Thanks!

---

<div class="post-metadata">

**Author:** ![jsocolar](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/jsocolar/32/2486_2.png) [@jsocolar](https://discourse.mc-stan.org/u/jsocolar)\
**Post date:** [May 23, 2023, 5:48pm UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262/9 "2023-05-23T17:48:34Z")

</div>

There is a memory spike post-fitting when reading in the csvs. If this causes memory pressure, it may cause the process to hang. If this is your problem, you might be able to make progress by:

- re-fitting the model in an environment with more RAM
- avoiding a refit by moving the output csvs to an environment with more RAM and reading them there
- Reading the output csvs one at a time

---

<div class="post-metadata">

**Author:** ![joshuakrobertson](https://avatars.discourse-cdn.com/v4/letter/j/85e7bf/32.png) [@joshuakrobertson](https://discourse.mc-stan.org/u/joshuakrobertson)\
**Post date:** [August 11, 2023, 3:38pm UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262/10 "2023-08-11T15:38:18Z")

</div>

Thanks @jsocolar. I tried freeing memory and allocating more to R with no luck. However, execution in terminal (via Radian) works just fine. More evidence that Rstudio is a bit bulkier than necessary…

---

<div class="post-metadata">

**Author:** ![nico.sp](https://avatars.discourse-cdn.com/v4/letter/n/7ea924/32.png) [@nico.sp](https://discourse.mc-stan.org/u/nico.sp)\
**Post date:** [April 10, 2025, 11:12am UTC](https://discourse.mc-stan.org/t/brms-hangs-after-drawing-samples/26262/11 "2025-04-10T11:12:40Z")

</div>

I’m not sure if anyone still experiences this issue, but I recently encountered the same problem and managed to resolve it — unless I’m misunderstanding something. I wasn’t able to reproduce the issue using matti’s example (even with `N = 50000`) because the problem didn’t occur there. For context, I used the **rstan** backend, as I haven’t tried **cmdstanr** yet.

My original plan was to run the model and immediately view its output with `summary()`, like this:

```r
fit <- brm(
  formula = formula,
  data = data,
  family = family,
  file = file,
  cores = 4
)
summary(fit)

```

This approach worked perfectly for simpler models. However, with more complex models I observed that, after a while, all chains reached 100% completion but the expected summary output never appeared - even after hours. Additionally, the model wasn’t saved via `brm()`’s internal save functionality. Although CPU and RAM usage spiked during the model run, they eventually returned to baseline, indicating that this wasn’t an issue with system memory.

I tried several things, from restarting my computer to reducing the number of cores, but nothing resolved the issue. To determine whether the problem was with the model run or with `summary()`, I added a `print("Finished.")` statement immediately after the `brm()` call and ran the code only up to that point.

It turns out that this simple modification reliably resolved the issue for me:

```r
fit <- brm(
  formula = formula,
  data = data,
  family = family,
  file = file,
  cores = 4
)
print("Finished.")
# Wait until the model run is complete.
summary(fit)

```

Since applying this change, the print message `“Finished.”` appears every time and the model is correctly saved by `brm()`’s internal mechanism. Subsequently, using `summary()` works without any issues.

I hope this helps anyone encountering this problem.