check_model() executes but produces no plot

[OneGeLiu]

The check_model() is very useful function for my work. However, it has recently failed to work recently (only blank plots are displayed). I'm sure that both the 'performance' package and its dependencies are updated to the latest versions, but the issue persists. I'd appreciate any insight into this problem. Please let me know if you have any suggestions for a solution.

[bwiernik]

Can you give a reproducible example ?

[OneGeLiu]

Can you give a reproducible example ?

I tried running the following codes on another computer, where it successfully generate a plot:

library(performance)
lm <- lm(mpg ~ wt + cyl, data = mtcars)
check_model(lm)

However, on my computer, it only produced a blank plot. Therfore, I futher executed the following code:

p <- check_model(lm)
p # no plot on my computer
str(p)

#List of 8
 $ VIF        :Classes ‘check_collinearity’, ‘see_check_collinearity’ and 'data.frame':	2 obs. of  3 variables:
  ..$ x    : chr [1:2] "wt" "cyl"
  ..$ y    : num [1:2] 2.58 2.58
  ..$ group: chr [1:2] "low" "low"
  ..- attr(*, "CI")='data.frame':	2 obs. of  4 variables:
  .. ..$ VIF_CI_low       : num [1:2] 1.74 1.74
  .. ..$ VIF_CI_high      : num [1:2] 4.39 4.39
  .. ..$ Tolerance_CI_low : num [1:2] 0.228 0.228
  .. ..$ Tolerance_CI_high: num [1:2] 0.576 0.576
 $ QQ         :'data.frame':	32 obs. of  2 variables:
  ..$ x: num [1:32] 10.3 10.6 10.9 14.6 15.4 ...
  ..$ y: num [1:32] -1.812 -1.426 -1.414 -0.832 -0.83 ...
 $ NORM       :'data.frame':	1024 obs. of  3 variables:
  ..$ x    : num [1:1024] -4.29 -4.28 -4.27 -4.26 -4.25 ...
  ..$ y    : num [1:1024] 0.034 0.0343 0.0345 0.0348 0.0351 ...
  ..$ curve: num [1:1024] 0.0361 0.0364 0.0367 0.0369 0.0372 ...
 $ NCV        :'data.frame':	32 obs. of  2 variables:
  ..$ x: num [1:32] 22.3 21.5 26.3 20.4 16.6 ...
  ..$ y: num [1:32] -1.279 -0.465 -3.452 1.019 2.053 ...
 $ HOMOGENEITY:'data.frame':	32 obs. of  2 variables:
  ..$ x: num [1:32] 22.3 21.5 26.3 20.4 16.6 ...
  ..$ y: num [1:32] 0.716 0.43 1.184 0.635 0.916 ...
 $ OUTLIERS   : 'check_outliers' logi [1:32] FALSE FALSE FALSE FALSE FALSE FALSE ...
  ..- attr(*, "data")='data.frame':	32 obs. of  4 variables:
  .. ..$ Row          : int [1:32] 1 2 3 4 5 6 7 8 9 10 ...
  .. ..$ Distance_Cook: num [1:32] 0.005077 0.000444 0.056776 0.001803 0.023527 ...
  .. ..$ Outlier_Cook : num [1:32] 0 0 0 0 0 0 0 0 0 0 ...
  .. ..$ Outlier      : num [1:32] 0 0 0 0 0 0 0 0 0 0 ...
  ..- attr(*, "threshold")=List of 1
  .. ..$ cook: num 0.808
  ..- attr(*, "method")= chr "cook"
  ..- attr(*, "text_size")= num 3
  ..- attr(*, "influential_obs")='data.frame':	32 obs. of  7 variables:
  .. ..$ Hat           : num [1:32] 0.0548 0.0376 0.0798 0.0321 0.0912 ...
  .. ..$ Cooks_Distance: num [1:32] 0.005077 0.000444 0.056776 0.001803 0.023527 ...
  .. ..$ Predicted     : num [1:32] 22.3 21.5 26.3 20.4 16.6 ...
  .. ..$ Residuals     : num [1:32] -1.279 -0.465 -3.452 1.019 2.053 ...
  .. ..$ Std_Residuals : num [1:32] -0.512 -0.185 -1.402 0.404 0.839 ...
  .. ..$ Index         : int [1:32] 1 2 3 4 5 6 7 8 9 10 ...
  .. ..$ Influential   : chr [1:32] "OK" "OK" "OK" "OK" ...
  .. ..- attr(*, "cook_levels")= num 0.71
  .. ..- attr(*, "n_params")= int 3
  ..- attr(*, "variables")= chr "(Whole model)"
  ..- attr(*, "raw_data")='data.frame':	32 obs. of  3 variables:
  .. ..$ mpg: num [1:32] 21 21 22.8 21.4 18.7 18.1 14.3 24.4 22.8 19.2 ...
  .. ..$ wt : num [1:32] 2.62 2.88 2.32 3.21 3.44 ...
  .. ..$ cyl: num [1:32] 6 6 4 6 8 6 8 4 4 6 ...
  ..- attr(*, "outlier_var")= list()
  ..- attr(*, "outlier_count")=List of 2
  .. ..$ cook:'data.frame':	0 obs. of  1 variable:
  .. .. ..$ Row: int(0) 
  .. ..$ all :'data.frame':	0 obs. of  1 variable:
  .. .. ..$ Row: num(0) 
 $ INFLUENTIAL:'data.frame':	32 obs. of  7 variables:
  ..$ Hat           : num [1:32] 0.0548 0.0376 0.0798 0.0321 0.0912 ...
  ..$ Cooks_Distance: num [1:32] 0.005077 0.000444 0.056776 0.001803 0.023527 ...
  ..$ Predicted     : num [1:32] 22.3 21.5 26.3 20.4 16.6 ...
  ..$ Residuals     : num [1:32] -1.279 -0.465 -3.452 1.019 2.053 ...
  ..$ Std_Residuals : num [1:32] -0.512 -0.185 -1.402 0.404 0.839 ...
  ..$ Index         : int [1:32] 1 2 3 4 5 6 7 8 9 10 ...
  ..$ Influential   : chr [1:32] "OK" "OK" "OK" "OK" ...
  ..- attr(*, "cook_levels")= num 0.71
  ..- attr(*, "n_params")= int 3
 $ PP_CHECK   :Classes ‘performance_pp_check’, ‘see_performance_pp_check’ and 'data.frame':	32 obs. of  51 variables:
  ..$ sim_1 : num [1:32] 23.3 17.4 27.5 17.6 15 ...
  ..$ sim_2 : num [1:32] 22.9 22 29.1 18.3 16.4 ...
  ..$ sim_3 : num [1:32] 23.6 20.9 24.7 18.3 18.8 ...
  ..$ sim_4 : num [1:32] 23 23.9 28.9 19.6 17.8 ...
  ..$ sim_5 : num [1:32] 20.7 24.3 22.3 20.7 16.7 ...
  ..$ sim_6 : num [1:32] 24.5 22.6 23 20 17.7 ...
  ..$ sim_7 : num [1:32] 26.8 17.8 22.4 23.7 14.8 ...
  ..$ sim_8 : num [1:32] 22.5 19.4 27.2 20.6 15.4 ...
  ..$ sim_9 : num [1:32] 20.9 20.9 28.8 19 15.3 ...
  ..$ sim_10: num [1:32] 20.9 25.5 30.8 18.5 19.4 ...
  ..$ sim_11: num [1:32] 19.6 20.8 23.5 25.8 19.3 ...
  ..$ sim_12: num [1:32] 19.1 19.4 25.1 20.7 17.1 ...
  ..$ sim_13: num [1:32] 24.3 19.7 30.9 21.5 15.3 ...
  ..$ sim_14: num [1:32] 26 22.3 23.5 20.5 19.1 ...
  ..$ sim_15: num [1:32] 25.1 21 27.9 17.9 13.8 ...
  ..$ sim_16: num [1:32] 18.8 26 27.7 18.9 14 ...
  ..$ sim_17: num [1:32] 27.4 20 25.5 26.9 14.9 ...
  ..$ sim_18: num [1:32] 23 21.6 23.1 19.2 17.8 ...
  ..$ sim_19: num [1:32] 19.9 22.8 28.4 22.3 12.7 ...
  ..$ sim_20: num [1:32] 23.6 22 23.8 15.1 15.5 ...
  ..$ sim_21: num [1:32] 19.3 16.6 23.6 18.7 18.5 ...
  ..$ sim_22: num [1:32] 18.7 26 29.3 17.9 18.5 ...
  ..$ sim_23: num [1:32] 19.2 19.4 23 21.3 17.8 ...
  ..$ sim_24: num [1:32] 20.7 18.7 31.7 17.6 18.7 ...
  ..$ sim_25: num [1:32] 21.3 22.1 16.9 16.9 17.8 ...
  ..$ sim_26: num [1:32] 23.1 17.9 25.1 18.2 11.6 ...
  ..$ sim_27: num [1:32] 19.5 22.9 27.3 18.8 14.4 ...
  ..$ sim_28: num [1:32] 21.3 19.4 26 16.4 11.9 ...
  ..$ sim_29: num [1:32] 21.4 21.6 25.1 18.4 16.2 ...
  ..$ sim_30: num [1:32] 21.7 22 27.4 16.6 19.7 ...
  ..$ sim_31: num [1:32] 22.4 19.7 25.7 18.7 13.1 ...
  ..$ sim_32: num [1:32] 16.3 22.4 23.1 22.1 15.9 ...
  ..$ sim_33: num [1:32] 22.2 22.4 27.5 24.7 19.6 ...
  ..$ sim_34: num [1:32] 22.2 21.9 31.9 21.9 15.1 ...
  ..$ sim_35: num [1:32] 20.4 23.8 22.6 19.2 14.8 ...
  ..$ sim_36: num [1:32] 24.3 20.1 27 19.3 18.6 ...
  ..$ sim_37: num [1:32] 21.3 16.7 25.9 13.9 19.2 ...
  ..$ sim_38: num [1:32] 22.7 20 28.7 24 16.6 ...
  ..$ sim_39: num [1:32] 21.8 28.7 25.5 18.3 15.9 ...
  ..$ sim_40: num [1:32] 20.4 21 25.9 21.2 15.7 ...
  ..$ sim_41: num [1:32] 20.7 20.4 29.5 20.6 16.6 ...
  ..$ sim_42: num [1:32] 22.6 22.6 29.5 20.8 15.7 ...
  ..$ sim_43: num [1:32] 23.3 24 26 25.4 15.3 ...
  ..$ sim_44: num [1:32] 23.3 19.1 30 19.4 16.3 ...
  ..$ sim_45: num [1:32] 22.4 21.6 25.4 19.3 17.7 ...
  ..$ sim_46: num [1:32] 20.1 21.2 29.8 22.3 15.4 ...
  ..$ sim_47: num [1:32] 23.6 17.5 25.1 20.2 18.1 ...
  ..$ sim_48: num [1:32] 21.9 19.9 21.6 21 17.4 ...
  ..$ sim_49: num [1:32] 24.2 26.5 26.6 19.6 17.5 ...
  ..$ sim_50: num [1:32] 21.2 24.8 27.8 18.7 13.4 ...
  ..$ y     : num [1:32] 21 21 22.8 21.4 18.7 18.1 14.3 24.4 22.8 19.2 ...
  ..- attr(*, "seed")= int [1:626] 10403 81 1750193880 1883868742 497830313 587170445 1634295632 -1171331099 744345694 -872982424 ...
  ..- attr(*, "check_range")= logi FALSE
  ..- attr(*, "response_name")= chr "mpg"
  ..- attr(*, "bandwidth")= chr "nrd"
  ..- attr(*, "model_info")=List of 51
  .. ..$ is_binomial     : logi FALSE
  .. ..$ is_bernoulli    : logi FALSE
  .. ..$ is_count        : logi FALSE
  .. ..$ is_poisson      : logi FALSE
  .. ..$ is_negbin       : logi FALSE
  .. ..$ is_beta         : logi FALSE
  .. ..$ is_betabinomial : logi FALSE
  .. ..$ is_orderedbeta  : logi FALSE
  .. ..$ is_dirichlet    : logi FALSE
  .. ..$ is_exponential  : logi FALSE
  .. ..$ is_logit        : logi FALSE
  .. ..$ is_probit       : logi FALSE
  .. ..$ is_censored     : logi FALSE
  .. ..$ is_truncated    : logi FALSE
  .. ..$ is_survival     : logi FALSE
  .. ..$ is_linear       : logi TRUE
  .. ..$ is_tweedie      : logi FALSE
  .. ..$ is_zeroinf      : logi FALSE
  .. ..$ is_zero_inflated: logi FALSE
  .. ..$ is_dispersion   : logi FALSE
  .. ..$ is_hurdle       : logi FALSE
  .. ..$ is_ordinal      : logi FALSE
  .. ..$ is_cumulative   : logi FALSE
  .. ..$ is_multinomial  : logi FALSE
  .. ..$ is_categorical  : logi FALSE
  .. ..$ is_mixed        : logi FALSE
  .. ..$ is_multivariate : logi FALSE
  .. ..$ is_trial        : logi FALSE
  .. ..$ is_bayesian     : logi FALSE
  .. ..$ is_gam          : logi FALSE
  .. ..$ is_anova        : logi FALSE
  .. ..$ is_timeseries   : logi FALSE
  .. ..$ is_ttest        : logi FALSE
  .. ..$ is_correlation  : logi FALSE
  .. ..$ is_onewaytest   : logi FALSE
  .. ..$ is_chi2test     : logi FALSE
  .. ..$ is_ranktest     : logi FALSE
  .. ..$ is_levenetest   : logi FALSE
  .. ..$ is_variancetest : logi FALSE
  .. ..$ is_xtab         : logi FALSE
  .. ..$ is_proptest     : logi FALSE
  .. ..$ is_binomtest    : logi FALSE
  .. ..$ is_ftest        : logi FALSE
  .. ..$ is_meta         : logi FALSE
  .. ..$ is_wiener       : logi FALSE
  .. ..$ is_rtchoice     : logi FALSE
  .. ..$ is_mixture      : logi FALSE
  .. ..$ link_function   : chr "identity"
  .. ..$ family          : chr "gaussian"
  .. ..$ n_obs           : int 32
  .. ..$ n_grouplevels   : NULL
  ..- attr(*, "type")= chr "density"
 - attr(*, "class")= chr [1:2] "check_model" "see_check_model"
 - attr(*, "panel")= logi TRUE
 - attr(*, "dot_size")= num 2
 - attr(*, "line_size")= num 0.8
 - attr(*, "base_size")= num 10
 - attr(*, "axis_title_size")= num 10
 - attr(*, "title_size")= num 12
 - attr(*, "check")= chr "all"
 - attr(*, "alpha")= num 0.2
 - attr(*, "dot_alpha")= num 0.8
 - attr(*, "show_dots")= logi TRUE
 - attr(*, "detrend")= logi TRUE
 - attr(*, "colors")= chr [1:3] "#3aaf85" "#1b6ca8" "#cd201f"
 - attr(*, "theme")= chr "see::theme_lucid"
 - attr(*, "model_info")=List of 51
  ..$ is_binomial     : logi FALSE
  ..$ is_bernoulli    : logi FALSE
  ..$ is_count        : logi FALSE
  ..$ is_poisson      : logi FALSE
  ..$ is_negbin       : logi FALSE
  ..$ is_beta         : logi FALSE
  ..$ is_betabinomial : logi FALSE
  ..$ is_orderedbeta  : logi FALSE
  ..$ is_dirichlet    : logi FALSE
  ..$ is_exponential  : logi FALSE
  ..$ is_logit        : logi FALSE
  ..$ is_probit       : logi FALSE
  ..$ is_censored     : logi FALSE
  ..$ is_truncated    : logi FALSE
  ..$ is_survival     : logi FALSE
  ..$ is_linear       : logi TRUE
  ..$ is_tweedie      : logi FALSE
  ..$ is_zeroinf      : logi FALSE
  ..$ is_zero_inflated: logi FALSE
  ..$ is_dispersion   : logi FALSE
  ..$ is_hurdle       : logi FALSE
  ..$ is_ordinal      : logi FALSE
  ..$ is_cumulative   : logi FALSE
  ..$ is_multinomial  : logi FALSE
  ..$ is_categorical  : logi FALSE
  ..$ is_mixed        : logi FALSE
  ..$ is_multivariate : logi FALSE
  ..$ is_trial        : logi FALSE
  ..$ is_bayesian     : logi FALSE
  ..$ is_gam          : logi FALSE
  ..$ is_anova        : logi FALSE
  ..$ is_timeseries   : logi FALSE
  ..$ is_ttest        : logi FALSE
  ..$ is_correlation  : logi FALSE
  ..$ is_onewaytest   : logi FALSE
  ..$ is_chi2test     : logi FALSE
  ..$ is_ranktest     : logi FALSE
  ..$ is_levenetest   : logi FALSE
  ..$ is_variancetest : logi FALSE
  ..$ is_xtab         : logi FALSE
  ..$ is_proptest     : logi FALSE
  ..$ is_binomtest    : logi FALSE
  ..$ is_ftest        : logi FALSE
  ..$ is_meta         : logi FALSE
  ..$ is_wiener       : logi FALSE
  ..$ is_rtchoice     : logi FALSE
  ..$ is_mixture      : logi FALSE
  ..$ link_function   : chr "identity"
  ..$ family          : chr "gaussian"
  ..$ n_obs           : int 32
  ..$ n_grouplevels   : NULL
 - attr(*, "bandwidth")= chr "nrd"
 - attr(*, "type")= chr "density"
 - attr(*, "model_class")= chr "lm"`

I found that althou a list was retured, the plot still faied to render

Since I dont know how to reproduce this bug on other computers, I am providing my R information below to help diagnose the problem:

sessionInfo()
#R version 4.5.1 (2025-06-13 ucrt)
Platform: x86_64-w64-mingw32/x64
Running under: Windows 11 x64 (build 26200)

Matrix products: default
  LAPACK version 3.12.1

locale:
[1] LC_COLLATE=Chinese (Simplified)_China.utf8  LC_CTYPE=Chinese (Simplified)_China.utf8    LC_MONETARY=Chinese (Simplified)_China.utf8
[4] LC_NUMERIC=C                                LC_TIME=Chinese (Simplified)_China.utf8    

time zone: Asia/Hong_Kong
tzcode source: internal

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

other attached packages:
[1] see_0.12.0.1         performance_0.15.2.3

loaded via a namespace (and not attached):
 [1] generics_0.1.4          robustbase_0.99-4-1     qqconf_1.3.2            bitops_1.0-9            lattice_0.22-7         
 [6] pracma_2.4.4            lme4_1.1-37             magrittr_2.0.3          caTools_1.18.3          grid_4.5.1             
[11] RColorBrewer_1.1-3      iterators_1.0.14        pkgload_1.4.0           doParallel_1.0.17       foreach_1.5.2          
[16] Matrix_1.7-3            opdisDownsampling_1.0.1 ggrepel_0.9.6           mgcv_1.9-3              scales_1.4.0           
[21] codetools_0.2-20        reformulas_0.4.0        Rdpack_2.6.4            cli_3.6.4               pbmcapply_1.5.1        
[26] rlang_1.1.6             rbibutils_2.3           splines_4.5.1           withr_3.0.2             tools_4.5.1            
[31] datawizard_1.3.0        parallel_4.5.1          nloptr_2.2.1            minqa_1.2.8             dplyr_1.1.4            
[36] ggplot2_4.0.0           bayestestR_0.17.0       boot_1.3-31             vctrs_0.6.5             R6_2.6.1               
[41] lifecycle_1.0.4         MASS_7.3-65             insight_1.4.2           pkgconfig_2.0.3         pillar_1.11.1          
[46] gtable_0.3.6            glue_1.8.0              Rcpp_1.0.14             DEoptimR_1.1-4          tibble_3.2.1           
[51] tidyselect_1.2.1        rstudioapi_0.17.1       farver_2.1.2            patchwork_1.3.2         nlme_3.1-168           
[56] labeling_0.4.3          qqplotr_0.0.6           DHARMa_0.4.7            compiler_4.5.1          S7_0.2.0               
[61] twosamples_2.0.1       `

I guess this might be a package conflict?

[bwiernik]

What is your version of RStudio and your screen resolution? Does the plot render if you click the View button in the plot pane to pop it out into a separate window?

[OneGeLiu]

What is your version of RStudio and your screen resolution? Does the plot render if you click the View button in the plot pane to pop it out into a separate window?

My RStudio version is RStudio 2025.09.2+418 .
My screen resolution is 1920*1080
I used the dev.new() to generate a separate window, but also generated a blank plot.

[bwiernik]

Can you check outside of RStudio (just running R GUI)

[OneGeLiu]

Can you check outside of RStudio (just running R GUI)

sorry, it still doesn't work

[bwiernik]

That's helpful for narrowing down the source

[hnairways]

Also have this problem, I split the checkmodel() output figures and found this error on QQ plot, Error in

qqplotr::stat_qq_band():
! Problem while converting geom to grob.
ℹ Error occurred in the 1st layer.
Caused by error in check_linewidth():
! could not find function "check_linewidth"
Run rlang::last_trace() to see where the error occurred.

And this error making a null plot at the end unless you generate plots one by one.
Hope that might give you some idea? @bwiernik
Thanks for the efforts!

[bljm]

Same issue, the problem is only with the qq plot library(performance)
model1 <- glmmTMB::glmmTMB(
count ~ mined + spp + (1 | site),
family = poisson,
data = glmmTMB::Salamanders
)
check_model(model1, check = c("vif","qq","reqq","homogeneity","pp_check","overdispersion")) # doesnt work
check_model(model1, check = c("vif","reqq","homogeneity","pp_check","overdispersion")) # works I run it in a Mac with updated glmmtb 1.1.13 version

[dana-and-monsters]

I have the same issue and I believe it started occurring after the recent updates to performance so I rolled back my performance version from 0.15.3 to 0.15.0 and then to 0.14.0 but no luck getting check_model to work again.

In some previous version updates I see there were some issues that someone suggested fixing with
performance::check_model(groove_max_3D_feasible, panel = FALSE) |> plot()
When I run that I get 4 plots that run:

$PP_CHECK

$NCV

$HOMOGENEITY

$OUTLIERS

And presumably when it tries to execute for the qqplot,
$QQ

Then:
Error in qqplotr::stat_qq_band():
! Problem while converting geom to grob.
ℹ Error occurred in the 1st layer.
Caused by error in check_linewidth():
! could not find function "check_linewidth"
Run rlang::last_trace() to see where the error occurred.

My qqplotr says version 0.0.6 though their github and webpage suggest the last version was 0.0.5 and released in 2020. It looks like they are trying to deploy 0.1.0 since September so maybe something they did caused a down-end bug here?

[strengejacke]

@OneGeLiu were there any messages from check_model()? The issue of missing plot may occur when you increased the screen scaling in the Windows settings.

[OneGeLiu]

@OneGeLiu were there any messages from check_model()? The issue of missing plot may occur when you increased the screen scaling in the Windows settings.
I’m glad to report that my check_model() function has recently started generating plots as expected again, though I’m still unsure what exactly triggered this recovery. Prior to this, when I executed the function, no messages were displayed. I have not made any changes to my screen resolution or screen scaling seetings lately. The only change I made to my R environment was installing several additional plotting packages. Thanks for your hepful solution!