Shinystan launch error: Unused Argument

Hello,

I am new to shinystan so apologies if I don’t provide enough info/format wrong or something, I can fix as required.

I am having a problem when launching shinystan for my model & after some forum searching I found nothing definite that would help.

After training model
by_segment_model_dated <- brm(…)

I attempt to launch shinystan:
`
Launching ShinyStan interface… for large models this may take some time.

Listening on http://127.0.0.1:6690
Warning: The size argument of element_line() is deprecated as of ggplot2 3.4.0.
ℹ Please use the linewidth argument instead.
This warning is displayed once every 8 hours.
Call lifecycle::last_lifecycle_warnings() to see where this warning was generated.
Warning: The select input “diagnostic_param” contains a large number of options; consider using server-side selectize for massively improved performance. See the Details section of the ?selectizeInput help topic.
Warning: The select input “ac_params” contains a large number of options; consider using server-side selectize for massively improved performance. See the Details section of the ?selectizeInput help topic.
Warning in .approxfun(x, y, v, method, yleft, yright, f, na.rm) :
NAs introduced by coercion
Warning in .approxfun(x, y, v, method, yleft, yright, f, na.rm) :
NAs introduced by coercion
Warning in .approxfun(x, y, v, method, yleft, yright, f, na.rm) :
NAs introduced by coercion
Warning: Error in p: unused argument (em(“posterior predictive distribution”))
125: dots_list
124: div
103: source_ui [ui_utils.R#2]
86: source_ui [ui_utils.R#2]
6: shiny::runApp
5: launch
4: launch_shinystan.shinystan
2: launch_shinystan.brmsfit
1: launch_shinystan
`

The shinystan window opens with the below error message:

unused argument(en(“posterior predictive distribution”))

Shinystan used to launch for my colleague with these models so I don’t think the problem lies in model code.

I don’t really know where to start looking, so any suggestions would be welcome!

Here’s my R session info (didn’t bother including loaded via a namespace):

R version 4.2.1 (2022-06-23 ucrt)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 19042)

Matrix products: default

locale:
[1] LC_COLLATE=English_United States.utf8  LC_CTYPE=English_United States.utf8    LC_MONETARY=English_United States.utf8
[4] LC_NUMERIC=C                           LC_TIME=English_United States.utf8    

attached base packages:
[1] parallel  stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] viridis_0.6.3         viridisLite_0.4.2     forcats_1.0.0         purrr_1.0.1           readr_2.1.4          
 [6] tibble_3.2.1          tidyverse_2.0.0       tidyr_1.3.0           tidybayes_3.0.4       stringr_1.5.0        
[11] skimr_2.1.5           shinystan_2.6.0       shiny_1.7.4           rstanarm_2.21.4       rstan_2.21.8         
[16] StanHeaders_2.21.0-7  RODBC_1.3-20          rjson_0.2.21          rAzureBatch_0.7.0     pracma_2.4.2         
[21] plotly_4.10.1         patchwork_1.1.2       Microsoft365R_2.3.4   microbenchmark_1.4.10 mice_3.15.0          
[26] magrittr_2.0.3        lubridate_1.9.3       htmlwidgets_1.6.2     Hmisc_5.0-1           gtExtras_0.4.5       
[31] gt_0.9.0              ggplot2_3.4.2         dplyr_1.1.2           doParallel_1.0.17     iterators_1.0.14     
[36] foreach_1.5.2         devtools_2.4.5        usethis_2.1.6         data.table_1.14.8     conflicted_1.2.0     
[41] cmdstanr_0.5.3        brms_2.19.0           Rcpp_1.0.10           blastula_0.3.3        bayesplot_1.10.0     
[46] AzureStor_3.7.0       arrow_11.0.0.3

OK after a brute-force investigation the culprit turned out to be a piece of code in the process which assigns ecdf() to a variable named p:

p ← ecdf(k)

Clearly somewhere in stan process there is a variable named p, which this reassigns, during the shinystan launch.

Once the variable named p is removed from the session environment (I renamed it for the ecdf() assignment) shinystan launches as expected.

Thanks for your messages. I got the same error and I could resolved it thanks to you. Such a random error. : )

Thanks! It’s really nice to know this was helpful.