Robyn is an experimental, AI/ML-powered and open sourced Marketing Mix Modeling (MMM) package from Meta Marketing Science. Our mission is to democratise modeling knowledge, inspire the industry through innovation, reduce human bias in the modeling process & build a strong open source marketing science community.
Whenever I try to run the model, I get a connection error after running the first trial. I saw a previous post on this regarding trying to lower the hyperparameters or number of trials and cores used, but this does nothing for me. I understand the last post also mentioned this was a memory issue but even after entirely restarting RStudio, emptying my environment, and reducing all memory load issues, it still does not run.
The model managed to run once, unfortunately this was when it was not an optimal model yet but only a trial. Since then, it has never run successfully.
update*: I have also tried to run on geometric now with only 56 hyperparameters and still does not run. It always stops after running one trial, no matter how many iterations. I also noticed that although I force the multicore usage, it still continues to fallback to 1 core.
'Input data has 105 weeks in total: 2021-01-04 to 2023-01-02
Initial model is built on rolling window of 105 week: 2021-01-04 to 2023-01-02
Time-series validation with train_size range of 50%-80% of the data...
Using weibull_pdf adstocking with 74 hyperparameters (74 to iterate + 0 fixed) on 1 core (Windows fallback)
Starting 5 trials with 500 iterations each using TwoPointsDE nevergrad algorithm...
Running trial 1 of 5
|=============================================================================================================================== | 99%Error in summary.connection(connection) : invalid connection
In addition: Warning message:
In check_iteration(InputCollect$calibration_input, iterations, trials, :
We recommend to run at least 2000 iterations per trial and 5 trials to build initial model`
Environment & Robyn version
> packageVersion("Robyn") [1] ‘3.9.0’
`R version 4.2.2 (2022-10-31 ucrt)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 19044)
Project Robyn
Describe issue
Whenever I try to run the model, I get a connection error after running the first trial. I saw a previous post on this regarding trying to lower the hyperparameters or number of trials and cores used, but this does nothing for me. I understand the last post also mentioned this was a memory issue but even after entirely restarting RStudio, emptying my environment, and reducing all memory load issues, it still does not run.
The model managed to run once, unfortunately this was when it was not an optimal model yet but only a trial. Since then, it has never run successfully.
update*: I have also tried to run on geometric now with only 56 hyperparameters and still does not run. It always stops after running one trial, no matter how many iterations. I also noticed that although I force the multicore usage, it still continues to fallback to 1 core.
'Input data has 105 weeks in total: 2021-01-04 to 2023-01-02 Initial model is built on rolling window of 105 week: 2021-01-04 to 2023-01-02 Time-series validation with train_size range of 50%-80% of the data... Using weibull_pdf adstocking with 74 hyperparameters (74 to iterate + 0 fixed) on 1 core (Windows fallback)
Environment & Robyn version
> packageVersion("Robyn") [1] ‘3.9.0’
`R version 4.2.2 (2022-10-31 ucrt) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows 10 x64 (build 19044)
Matrix products: default
locale: [1] LC_COLLATE=English_Netherlands.utf8 LC_CTYPE=English_Netherlands.utf8 LC_MONETARY=English_Netherlands.utf8 [4] LC_NUMERIC=C LC_TIME=English_Netherlands.utf8
attached base packages: [1] stats graphics grDevices utils datasets methods base
other attached packages: [1] readxl_1.4.1 doRNG_1.8.6 rngtools_1.5.2 foreach_1.5.2 Robyn_3.9.0 reticulate_1.28
loaded via a namespace (and not attached): [1] colorspace_2.0-3 ellipsis_0.3.2 ggridges_0.5.4 rprojroot_2.0.3 rstudioapi_0.14 farver_2.1.1
[7] rstan_2.21.8 remotes_2.4.2 fansi_1.0.3 lubridate_1.9.0 xml2_1.3.3 codetools_0.2-18
[13] splines_4.2.2 doParallel_1.0.17 jsonlite_1.8.4 nloptr_2.0.3 pROC_1.18.0 png_0.1-8
[19] compiler_4.2.2 httr_1.4.5 assertthat_0.2.1 Matrix_1.5-1 lazyeval_0.2.2 cli_3.6.0
[25] rPref_1.4.0 prettyunits_1.1.1 tools_4.2.2 igraph_1.4.2 gtable_0.3.3 glue_1.6.2
[31] dplyr_1.0.10 rappdirs_0.3.3 Rcpp_1.0.10 prophet_1.0 carData_3.0-5 cellranger_1.1.0
[37] fracdiff_1.5-2 h2o_3.40.0.1 vctrs_0.5.1 urca_1.3-3 nlme_3.1-160 iterators_1.0.14
[43] lmtest_0.9-40 timeDate_4022.108 stringr_1.5.0 ps_1.7.2 openxlsx_4.2.5.2 rvest_1.0.3
[49] timechange_0.2.0 lifecycle_1.0.3 zoo_1.8-11 scales_1.2.1 ragg_1.2.5 parallel_4.2.2
[55] inline_0.3.19 RColorBrewer_1.1-3 rpart.plot_3.1.1 yaml_2.3.6 quantmod_0.4.20 curl_5.0.0
[61] gridExtra_2.3 ggplot2_3.4.2 loo_2.6.0 StanHeaders_2.21.0-7 rpart_4.1.19 stringi_1.7.12
[67] tseries_0.10-53 desc_1.4.2 TTR_0.24.3 pkgbuild_1.4.0 zip_2.2.2 shape_1.4.6
[73] systemfonts_1.0.4 rlang_1.1.0 pkgconfig_2.0.3 bitops_1.0-7 matrixStats_0.63.0 lattice_0.20-45
[79] purrr_1.0.1 patchwork_1.1.2 labeling_0.4.2 processx_3.8.0 tidyselect_1.2.0 here_1.0.1
[85] plyr_1.8.8 magrittr_2.0.3 R6_2.5.1 generics_0.1.3 DBI_1.1.3 pillar_1.9.0
[91] withr_2.5.0 mgcv_1.8-41 xts_0.12.2 survival_3.4-0 abind_1.4-5 RCurl_1.98-1.12
[97] nnet_7.3-18 tibble_3.1.8 crayon_1.5.2 car_3.1-1 utf8_1.2.2 lares_5.2.1
[103] grid_4.2.2 minpack.lm_1.2-3 callr_3.7.3 forecast_8.20 digest_0.6.31 tidyr_1.2.1
[109] extraDistr_1.9.1 textshaping_0.3.6 RcppParallel_5.1.7 stats4_4.2.2 munsell_0.5.0 glmnet_4.1-7
[115] quadprog_1.5-8 `