Linking multiple mapping multiple dynamic targets to a target #1321

edgBR · 2020-09-11T15:32:51Z

Prework

Original workflow: #1293
Modified workflow to correct the DAG: #1294
First attemp to rewrite the plan using dynamic targets: #1311
Linking dynamic targets: #1314

Description

Dear colleagues, keeping working in my final drake workflow to fix my issues with the fable package.

I have now manage to use the dynamic targets to get my data for different set of ids and sensors. Besides the training works quite nicely (even that I am having some issues with the paralelization of the code, but probably I will not open anything until I dont do more ground work my self).

The workflow that is currently up and running is the following:

And is nicely combining cross and map in a way that I can see the training times for different algorithms.

However now I am trying to apply and accuracy function to evaluate the forecasts in the testing data and then, select the best model, refit it to the whole historical data and then forecast the future.

Reproducible example

My test plan looks as follows:

new_plan_dynamic_branch_test <- drake_plan(
  unit_metadata = getMetadata(
    environment = "PROD",
    key_directory = config_parameters$LOCAL_CONFIG$DirectoryKeyCloud_RStudio,
    operex_schema = config_parameters$SF_CONFIG$schema_name, db_src = c(1, 2, 3)
  ),
  distinct_b_numbers = c('xxxxxxxxxxxxx'),
  distinct_db_src = unit_metadata$db_src %>% unique() %>% as.numeric() %>% sort(),
  distinct_snsr_keys = getCountersKeys(
    environment = "PROD",
    key_directory = config_parameters$LOCAL_CONFIG$DirectoryKeyCloud_RStudio,
    operex_schema = config_parameters$SF_CONFIG$schema_name
  ),
  getIndividualData = target(
    getIndividualCounterData(
      environment = "PROD",
      key_directory = config_parameters$RSTUDIO_CLOUD_CONF$KeyDir,
      operex_schema = config_parameters$SF_CONFIG$schema_name,
      package_b_number = distinct_b_numbers,
      counter = distinct_snsr_keys,
      db_src = distinct_db_src, 
      max_forecasting_horizon = argument_parser$horizon
    ),
    dynamic = cross( # Use `dynamic =` instead of `transform =`
      distinct_b_numbers, # no tidy evaluation needed for dynamic branching
      distinct_snsr_keys,
      distinct_db_src
    )
  ),
  processingData = target(
    featureEngineering(
      raw_data = getIndividualData, 
      max_forecasting_horizon = argument_parser$horizon
    ),
    dynamic = map(getIndividualData)
  ),
  training_data = target(
    timeSeriesSplitter(input_data = processingData,
             max_forecast_horizon = argument_parser$horizon,
             type = "train"
  ),
  dynamic = map(processingData)
  ),
  testing_data = target(
    timeSeriesSplitter(input_data = processingData,
             max_forecast_horizon = argument_parser$horizon,
             type = "test"
  ),
  dynamic = map(processingData)
  ),
  models_training = target( ### split train/validation set
    trainModels(
      input_data = training_data,
      max_forecast_horizon = argument_parser$horizon, 
      model_type = model_types,
      max_multisession_cores = argument_parser$sessions
    ),
    dynamic = map(training_data), # dynamic branching
    transform = map(model_type = !!model_types)
  ),
  accuracy = target(
    accuracy_explorer(
      mode = "test",
      models = models_training,
      max_forecast_horizon = argument_parser$horizon,
      bucket = argument_parser$outputbucket,
      bucket_folder = "test_lifecounter",
      testing_data = testing_data
  ), 
  dynamic = map(models_training, testing_data)
  )
)

Desired result

Is there a way to dynamically map accuracy to both the individual steps of models_training and testing data?

Session info

R version 4.0.0 (2020-04-24)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 16.04.6 LTS

Matrix products: default
BLAS:   /usr/lib/atlas-base/atlas/libblas.so.3.0
LAPACK: /usr/lib/atlas-base/atlas/liblapack.so.3.0

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     LC_MONETARY=en_US.UTF-8   
 [6] LC_MESSAGES=en_US.UTF-8    LC_PAPER=en_US.UTF-8       LC_NAME=C                  LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

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

other attached packages:
 [1] rstudioapi_0.11     tidyselect_1.1.0    feather_0.3.5       xts_0.12-0          zoo_1.8-8           DescTools_0.99.35   data.table_1.12.8  
 [8] qs_0.23.2           furrr_0.1.0         future_1.17.0       imputeTS_3.0        timetk_2.0.0        feasts_0.1.5        forecast_8.12      
[15] fable.prophet_0.1.0 Rcpp_1.0.5          tsibble_0.9.1       forcats_0.5.0       stringr_1.4.0       purrr_0.3.4         readr_1.3.1        
[22] tibble_3.0.3        ggplot2_3.3.2       tidyverse_1.3.0     aws.s3_0.3.21       drake_7.12.2        tictoc_1.0          ConfigParser_1.0.0 
[29] R6_2.4.1            ini_0.3.1           DBI_1.1.0           odbc_1.2.2          lubridate_1.7.9     tidyr_1.1.0         dplyr_1.0.0        
[36] fable_0.2.1         fabletools_0.2.1    optparse_1.6.6     

loaded via a namespace (and not attached):
 [1] colorspace_1.4-1      ellipsis_0.3.1        class_7.3-17          base64enc_0.1-3       fs_1.4.1              forecastML_0.9.0      listenv_0.8.0        
 [8] farver_2.0.3          getopt_1.20.3         bit64_0.9-7           mvtnorm_1.1-0         prodlim_2019.11.13    fansi_0.4.1           xml2_1.3.2           
[15] codetools_0.2-16      splines_4.0.0         jsonlite_1.6.1        packrat_0.5.0         broom_0.5.6           anytime_0.3.8         dbplyr_1.4.3         
[22] compiler_4.0.0        httr_1.4.1            backports_1.1.6       assertthat_0.2.1      Matrix_1.2-18         cli_2.0.2             htmltools_0.5.0      
[29] visNetwork_2.0.9      prettyunits_1.1.1     tools_4.0.0           igraph_1.2.5          gtable_0.3.0          glue_1.4.2            cellranger_1.1.0     
[36] fracdiff_1.5-1        vctrs_0.3.4           urca_1.3-0            nlme_3.1-147          lmtest_0.9-37         timeDate_3043.102     gower_0.2.1          
[43] globals_0.12.5        rvest_0.3.5           lifecycle_0.2.0       MASS_7.3-51.6         scales_1.1.1          ipred_0.9-9           aws.ec2metadata_0.2.0
[50] hms_0.5.3             parallel_4.0.0        expm_0.999-4          yaml_2.2.1            quantmod_0.4.17       curl_4.3              aws.signature_0.5.2  
[57] rpart_4.1-15          stringi_1.4.6         tseries_0.10-47       TTR_0.23-6            filelock_1.0.2        boot_1.3-25           lava_1.6.7           
[64] storr_1.2.1           rlang_0.4.7           pkgconfig_2.0.3       distributional_0.2.0  lattice_0.20-41       htmlwidgets_1.5.1     stinepack_1.4        
[71] recipes_0.1.12        bit_1.1-15.2          magrittr_1.5          generics_0.0.2        base64url_1.4         txtq_0.2.0            pillar_1.4.6         
[78] haven_2.2.0           withr_2.2.0           survival_3.1-12       nnet_7.3-14           modelr_0.1.7          crayon_1.3.4          RApiSerialize_0.1.0  
[85] progress_1.2.2        grid_4.0.0            readxl_1.3.1          blob_1.2.1            reprex_0.3.0          digest_0.6.25         stringfish_0.12.1    
[92] munsell_0.5.0         quadprog_1.5-8

wlandau · 2020-09-11T16:35:19Z

If I understand correctly, I think all you need is a transform = map(models_training) in accuracy.

accuracy = target(
  accuracy_explorer(
    mode = "test",
    models = models_training,
    max_forecast_horizon = argument_parser$horizon,
    bucket = argument_parser$outputbucket,
    bucket_folder = "test_lifecounter",
    testing_data = testing_data
  ), 
  dynamic = map(models_training, testing_data),
  transform = map(models_training)
)

In general terms, drake allows you to statically map over targets where you previously used static branching.

library(drake)
plan <- drake_plan(
  x = seq_len(40),
  y = target(
    f(x),
    transform = map(f = c(fun1, fun2)),
    dynamic = map(x)
  ),
  z = target(
    f(y),
    transform = map(y, f, .id = f),
    dynamic = map(y)
  )
)

print(plan)
#> # A tibble: 5 x 3
#>   target command      dynamic    
#>   <chr>  <expr>       <expr>     
#> 1 x      seq_len(40)  NA         
#> 2 y_fun1 fun1(x)      map(x)     
#> 3 y_fun2 fun2(x)      map(x)     
#> 4 z_fun1 fun1(y_fun1) map(y_fun1)
#> 5 z_fun2 fun2(y_fun2) map(y_fun2)

plot(plan)

^{Created on 2020-09-11 by the reprex package (v0.3.0)}

edgBR · 2020-09-14T15:28:17Z

Hi @wlandau

Almost there:

But when I check on the logs:

[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_auto_arima_ca5c2865
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_auto_arima_a3aff9ad
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_auto_arima_2966b24a
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_auto_arima_4a31b2c1
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_auto_arima_0e3676c9
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_auto_arima_997d675e
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_auto_arima_134802c2
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_auto_arima_f10513d9
> subtarget models_training_auto_arima_3258576f
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
▶ dynamic models_training_theta
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_theta_ef8b2553
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_theta_d7e53524
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_theta_bd5ab443
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_theta_ca5c2865
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_theta_a3aff9ad
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_theta_2966b24a
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_theta_4a31b2c1
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_theta_0e3676c9
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_theta_997d675e
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_theta_134802c2
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_theta_f10513d9
> subtarget models_training_theta_3258576f
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
▶ dynamic models_training_tslm
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_tslm_ef8b2553
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_tslm_d7e53524
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_tslm_bd5ab443
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_tslm_ca5c2865
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_tslm_a3aff9ad
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_tslm_2966b24a
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_tslm_4a31b2c1
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_tslm_0e3676c9
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_tslm_997d675e
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_tslm_134802c2
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_tslm_f10513d9
> subtarget models_training_tslm_3258576f
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
▶ dynamic models_training_regression_with_arima_errors
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_regression_with_arima_errors_ef8b2553
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_regression_with_arima_errors_d7e53524
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_regression_with_arima_errors_bd5ab443
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_regression_with_arima_errors_ca5c2865
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_regression_with_arima_errors_a3aff9ad
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_regression_with_arima_errors_2966b24a
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_regression_with_arima_errors_4a31b2c1
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_regression_with_arima_errors_0e3676c9
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_regression_with_arima_errors_997d675e
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_regression_with_arima_errors_134802c2
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_regression_with_arima_errors_f10513d9
> subtarget models_training_regression_with_arima_errors_3258576f
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
▶ dynamic models_training_prophet_multiplicative
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_prophet_multiplicative_ef8b2553
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_prophet_multiplicative_d7e53524
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_prophet_multiplicative_bd5ab443
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_prophet_multiplicative_ca5c2865
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_prophet_multiplicative_a3aff9ad
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_prophet_multiplicative_2966b24a
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_prophet_multiplicative_4a31b2c1
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_prophet_multiplicative_0e3676c9
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_prophet_multiplicative_997d675e
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_prophet_multiplicative_134802c2
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget models_training_prophet_multiplicative_f10513d9
> subtarget models_training_prophet_multiplicative_3258576f
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
■ finalize testing_data
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
■ finalize models_training_pophet_additive
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 71 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
■ finalize models_training_arima_with_monthly_fourier_components
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
■ finalize models_training_arima_with_yearly_fourier_components
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
■ finalize models_training_auto_arima
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 70 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
■ finalize models_training_theta
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
■ finalize models_training_tslm
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
■ finalize models_training_regression_with_arima_errors
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 105 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
▶ dynamic accuracy_models_training_pophet_additive
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 1"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget accuracy_models_training_pophet_additive_3be0b39d
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 2"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget accuracy_models_training_pophet_additive_d7f64bf5
[1] "Training auto_arima models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                           
[2] "Training arima_with_yearly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3" 
[3] "Training pophet_additive models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                      
[4] "Training prophet_multiplicative models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"               
[5] "Training arima_with_monthly_fourier_components models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"
[6] "Training regression_with_arima_errors models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"         
[7] "Training tslm models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                 
[8] "Training theta models for forecasting horizon 52 and package XXXXXXX snsr_key 110 and db_src 3"                                
[1] "Using 12 sessions from as future::plan()"
> subtarget accuracy_models_training_pophet_additive_e15cdd72
> subtarget accuracy_models_training_pophet_additive_b2c3b9a9
> subtarget accuracy_models_training_pophet_additive_3a93d846
> subtarget accuracy_models_training_pophet_additive_6c575b0c
> subtarget accuracy_models_training_pophet_additive_4aaa518d
> subtarget accuracy_models_training_pophet_additive_27fbb875
> subtarget accuracy_models_training_pophet_additive_87b6ab76
> subtarget accuracy_models_training_pophet_additive_7bf5e2b8
> subtarget accuracy_models_training_pophet_additive_8eab10df
> subtarget accuracy_models_training_pophet_additive_c2fa857b
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 71"
▶ dynamic accuracy_models_training_arima_with_monthly_fourier_components
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 71"
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_eee30f0d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 71"
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_b145fedd
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 70"
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_7f1d8cc2
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 70"
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_19512cef
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 70"
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_dbd3a301
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 105"
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_a6337dab
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 105"
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_4aaa518d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 105"
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_27fbb875
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 110"
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_87b6ab76
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 110"
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_7d25e48c
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 110"
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_116ff838
> subtarget accuracy_models_training_arima_with_monthly_fourier_components_c2fa857b
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 71"
▶ dynamic accuracy_models_training_arima_with_yearly_fourier_components
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 71"
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_eee30f0d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 71"
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_b145fedd
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 70"
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_7f1d8cc2
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 70"
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_19512cef
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 70"
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_dbd3a301
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 105"
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_a6337dab
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 105"
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_4aaa518d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 105"
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_27fbb875
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 110"
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_87b6ab76
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 110"
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_7d25e48c
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 110"
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_116ff838
> subtarget accuracy_models_training_arima_with_yearly_fourier_components_c2fa857b
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 71"
▶ dynamic accuracy_models_training_auto_arima
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 71"
> subtarget accuracy_models_training_auto_arima_eee30f0d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 71"
> subtarget accuracy_models_training_auto_arima_b145fedd
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 70"
> subtarget accuracy_models_training_auto_arima_7f1d8cc2
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 70"
> subtarget accuracy_models_training_auto_arima_19512cef
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 70"
> subtarget accuracy_models_training_auto_arima_dbd3a301
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 105"
> subtarget accuracy_models_training_auto_arima_a6337dab
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 105"
> subtarget accuracy_models_training_auto_arima_4aaa518d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 105"
> subtarget accuracy_models_training_auto_arima_27fbb875
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 110"
> subtarget accuracy_models_training_auto_arima_87b6ab76
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 110"
> subtarget accuracy_models_training_auto_arima_7d25e48c
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 110"
> subtarget accuracy_models_training_auto_arima_116ff838
> subtarget accuracy_models_training_auto_arima_c2fa857b
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 71"
▶ dynamic accuracy_models_training_theta
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 71"
> subtarget accuracy_models_training_theta_eee30f0d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 71"
> subtarget accuracy_models_training_theta_b145fedd
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 70"
> subtarget accuracy_models_training_theta_7f1d8cc2
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 70"
> subtarget accuracy_models_training_theta_19512cef
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 70"
> subtarget accuracy_models_training_theta_dbd3a301
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 105"
> subtarget accuracy_models_training_theta_a6337dab
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 105"
> subtarget accuracy_models_training_theta_4aaa518d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 105"
> subtarget accuracy_models_training_theta_27fbb875
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 110"
> subtarget accuracy_models_training_theta_87b6ab76
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 110"
> subtarget accuracy_models_training_theta_7d25e48c
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 110"
> subtarget accuracy_models_training_theta_116ff838
> subtarget accuracy_models_training_theta_c2fa857b
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 71"
▶ dynamic accuracy_models_training_tslm
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 71"
> subtarget accuracy_models_training_tslm_eee30f0d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 71"
> subtarget accuracy_models_training_tslm_b145fedd
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 70"
> subtarget accuracy_models_training_tslm_7f1d8cc2
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 70"
> subtarget accuracy_models_training_tslm_19512cef
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 70"
> subtarget accuracy_models_training_tslm_dbd3a301
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 105"
> subtarget accuracy_models_training_tslm_a6337dab
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 105"
> subtarget accuracy_models_training_tslm_4aaa518d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 105"
> subtarget accuracy_models_training_tslm_27fbb875
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 110"
> subtarget accuracy_models_training_tslm_87b6ab76
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 110"
> subtarget accuracy_models_training_tslm_7d25e48c
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 110"
> subtarget accuracy_models_training_tslm_116ff838
> subtarget accuracy_models_training_tslm_c2fa857b
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 71"
▶ dynamic accuracy_models_training_regression_with_arima_errors
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 71"
> subtarget accuracy_models_training_regression_with_arima_errors_eee30f0d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 71"
> subtarget accuracy_models_training_regression_with_arima_errors_b145fedd
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 70"
> subtarget accuracy_models_training_regression_with_arima_errors_7f1d8cc2
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 70"
> subtarget accuracy_models_training_regression_with_arima_errors_19512cef
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 70"
> subtarget accuracy_models_training_regression_with_arima_errors_dbd3a301
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 105"
> subtarget accuracy_models_training_regression_with_arima_errors_a6337dab
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 105"
> subtarget accuracy_models_training_regression_with_arima_errors_4aaa518d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 105"
> subtarget accuracy_models_training_regression_with_arima_errors_27fbb875
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 110"
> subtarget accuracy_models_training_regression_with_arima_errors_87b6ab76
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 110"
> subtarget accuracy_models_training_regression_with_arima_errors_7d25e48c
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 110"
> subtarget accuracy_models_training_regression_with_arima_errors_116ff838
> subtarget accuracy_models_training_regression_with_arima_errors_c2fa857b
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 71"
■ finalize models_training_prophet_multiplicative
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 71"
■ finalize accuracy_models_training_pophet_additive
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 71"
■ finalize accuracy_models_training_arima_with_monthly_fourier_components
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 70"
■ finalize accuracy_models_training_arima_with_yearly_fourier_components
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 70"
■ finalize accuracy_models_training_auto_arima
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 70"
■ finalize accuracy_models_training_theta
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 105"
■ finalize accuracy_models_training_tslm
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 105"
▶ dynamic accuracy_models_training_prophet_multiplicative
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 105"
> subtarget accuracy_models_training_prophet_multiplicative_eee30f0d
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 110"
> subtarget accuracy_models_training_prophet_multiplicative_b145fedd
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 110"
> subtarget accuracy_models_training_prophet_multiplicative_7f1d8cc2
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 110"
> subtarget accuracy_models_training_prophet_multiplicative_19512cef
> subtarget accuracy_models_training_prophet_multiplicative_dbd3a301
> subtarget accuracy_models_training_prophet_multiplicative_a6337dab
> subtarget accuracy_models_training_prophet_multiplicative_4aaa518d
> subtarget accuracy_models_training_prophet_multiplicative_27fbb875
> subtarget accuracy_models_training_prophet_multiplicative_87b6ab76
> subtarget accuracy_models_training_prophet_multiplicative_7d25e48c
> subtarget accuracy_models_training_prophet_multiplicative_116ff838
> subtarget accuracy_models_training_prophet_multiplicative_c2fa857b
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 71"
■ finalize accuracy_models_training_regression_with_arima_errors
**[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 71"
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 71"
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 70"
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 70"
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 70"
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 105"
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 105"
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 105"
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 1 and snsr_key 110"
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 2 and snsr_key 110"
[1] "Saving accuracy metrics for auto_arima model for package_b_number XXXXXXX and db_src 3 and snsr_key 110"**
■ finalize accuracy_models_training_prophet_multiplicative

It seems that is not saving the right metrics.

My plan looks as follows:

new_plan_dynamic_branch_test <- drake_plan(
  unit_metadata = getMetadata(
    environment = "PROD",
    key_directory = config_parameters$LOCAL_CONFIG$DirectoryKeyCloud_RStudio,
    operex_schema = config_parameters$SF_CONFIG$schema_name, db_src = c(1, 2, 3)
  ),
  distinct_b_numbers = c('XXXXXXXXXXX'),
  distinct_db_src = unit_metadata$db_src %>% unique() %>% as.numeric() %>% sort(),
  distinct_snsr_keys = getCountersKeys(
    environment = "PROD",
    key_directory = config_parameters$LOCAL_CONFIG$DirectoryKeyCloud_RStudio,
    operex_schema = config_parameters$SF_CONFIG$schema_name
  ),
  getIndividualData = target(
    getIndividualCounterData(
      environment = "PROD",
      key_directory = config_parameters$RSTUDIO_CLOUD_CONF$KeyDir,
      operex_schema = config_parameters$SF_CONFIG$schema_name,
      package_b_number = distinct_b_numbers,
      counter = distinct_snsr_keys,
      db_src = distinct_db_src, 
      max_forecasting_horizon = argument_parser$horizon
    ),
    dynamic = cross( # Use `dynamic =` instead of `transform =`
      distinct_b_numbers, # no tidy evaluation needed for dynamic branching
      distinct_snsr_keys,
      distinct_db_src
    )
  ),
  processingData = target(
    featureEngineering(
      raw_data = getIndividualData, 
      max_forecasting_horizon = argument_parser$horizon
    ),
    dynamic = map(getIndividualData)
  ),
  training_data = target(
    timeSeriesSplitter(input_data = processingData,
                       max_forecast_horizon = argument_parser$horizon,
                       type = "train"
    ),
    dynamic = map(processingData)
  ),
  testing_data = target(
    timeSeriesSplitter(input_data = processingData,
                       max_forecast_horizon = argument_parser$horizon,
                       type = "test"
    ),
    dynamic = map(processingData)
  ),
  models_training = target( ### split train/validation set
    trainModels(
      input_data = training_data,
      max_forecast_horizon = argument_parser$horizon, 
      model_type = model_types,
      max_multisession_cores = argument_parser$sessions
    ),
    dynamic = map(training_data), # dynamic branching
    transform = map(model_type = !!model_types)
  ),
  accuracy = target(
    accuracy_explorer(
      mode = "test",
      models = models_training,
      max_forecast_horizon = argument_parser$horizon,
      bucket = argument_parser$outputbucket,
      bucket_folder = "/test_lifecounter2",
      testing_data = testing_data
  ), 
  dynamic = map(models_training, testing_data),
  transform = map(models_training)
  )
)

And the accuracy function:

#' Testing accuracy metrics for different timeseries models
#' @author 
#' @param mode A string to trigger different routines depending of the value ("train" or "test)
#' @param models A mable object containing multiple timeseries models
#' @param max_forecast_horizon An integer that is used to identify the granularity of the forecasts
#' @param bucket A string specifiying the bucket name to save the metrics
#' @param bucket_folder A string specifiying the folder within the bucket
#' @return A tibble object


accuracy_explorer <- function(mode, models, max_forecast_horizon, bucket, bucket_folder, testing_data=NULL) {
  if(mode == "train") {
    accuracy_metrics <- models %>% accuracy() %>% 
      select(c(package_b_number, 
               snsr_key, 
               db_src, 
               .model, 
               .type,
               RMSE,
               MAE,
               MAPE,
               MASE)) 
    
    print(paste0("Saving accuracy metrics for ", attributes(models)$model, " model for",
                 " package_b_number ",
                 unique(accuracy_metrics$package_b_number),
                 " and db_src ", unique(accuracy_metrics$db_src),
                 " and snsr_key ", unique(accuracy_metrics$snsr_key)))
  } else if(mode=="test"){
    forecast_data <- forecast(models, testing_data)
    accuracy_metrics <- accuracy(forecast_data, testing_data) %>% 
      select(c(package_b_number, 
               snsr_key, 
               db_src, 
               .model, 
               .type,
               RMSE,
               MAE,
               MAPE,
               MASE)) %>% 
      mutate(horizon = max_forecast_horizon) 
    print(paste0("Saving accuracy metrics for ", attributes(models)$model, " model for",
                 " package_b_number ",
                 unique(accuracy_metrics$package_b_number),
                 " and db_src ", unique(accuracy_metrics$db_src),
                 " and snsr_key ", unique(accuracy_metrics$snsr_key)))
  }
  s3write_using(accuracy_metrics, 
                write_csv,
                object = paste0(bucket_folder, 
                                "/",
                                unique(accuracy_metrics$package_b_number),
                                "/horizon_",
                                max_forecast_horizon,
                                "/snsr_",
                                unique(accuracy_metrics$snsr_key),
                                "/db_src_",
                                unique(accuracy_metrics$db_src),
                                "/",
                                unique(accuracy_metrics$.model), 
                                "_accuracy_metrics.csv"), 
                bucket = bucket)
}

It seems that is not getting properly the attributes from the mable. Any hints of how to make a debug test case for this? I am not certain if the problem is drake or the mable object itself.

BR
/Edgar

edgBR · 2020-09-14T15:32:37Z

Seems that after enabling warnings again I am getting the following:

warnings()
Warning messages:
  1: In if (model_type == "prophet_additive") { ... :
      the condition has length > 1 and only the first element will be used
    2: In if (model_type == "prophet_additive") { ... :
        the condition has length > 1 and only the first element will be used
      3: In if (model_type == "prophet_additive") { ... :
          the condition has length > 1 and only the first element will be used
        4: In if (model_type == "auto_arima") { ... :
            the condition has length > 1 and only the first element will be used
          5: In if (model_type == "prophet_additive") { ... :
              the condition has length > 1 and only the first element will be used
            6: In if (model_type == "prophet_additive") { ... :
                the condition has length > 1 and only the first element will be used
              7: In if (model_type == "auto_arima") { ... :
                  the condition has length > 1 and only the first element will be used
                8: In if (model_type == "prophet_additive") { ... :
                    the condition has length > 1 and only the first element will be used
                  9: In if (model_type == "auto_arima") { ... :
                      the condition has length > 1 and only the first element will be used
                    10: In if (model_type == "prophet_additive") { ... :
                        the condition has length > 1 and only the first element will be used
                      11: In if (model_type == "prophet_additive") { ... :
                          the condition has length > 1 and only the first element will be used
                        12: In if (model_type == "prophet_additive") { ... :
                            the condition has length > 1 and only the first element will be used
                          13: In if (model_type == "prophet_additive") { ... :
                              the condition has length > 1 and only the first element will be used
                            14: In if (model_type == "auto_arima") { ... :
                                the condition has length > 1 and only the first element will be used
                              15: In if (model_type == "prophet_additive") { ... :
                                  the condition has length > 1 and only the first element will be used
                                16: In if (model_type == "prophet_additive") { ... :
                                    the condition has length > 1 and only the first element will be used
                                  17: In if (model_type == "prophet_additive") { ... :
                                      the condition has length > 1 and only the first element will be used
                                    18: In if (model_type == "prophet_additive") { ... :
                                        the condition has length > 1 and only the first element will be used
                                      19: In if (model_type == "auto_arima") { ... :
                                          the condition has length > 1 and only the first element will be used
                                        20: In if (model_type == "auto_arima") { ... :
                                            the condition has length > 1 and only the first element will be used
                                          21: In if (model_type == "prophet_additive") { ... :
                                              the condition has length > 1 and only the first element will be used
                                            22: In if (model_type == "prophet_additive") { ... :
                                                the condition has length > 1 and only the first element will be used
                                              23: In if (model_type == "prophet_additive") { ... :
                                                  the condition has length > 1 and only the first element will be used
                                                24: In if (model_type == "auto_arima") { ... :
                                                    the condition has length > 1 and only the first element will be used
                                                  25: In if (model_type == "auto_arima") { ... :
                                                      the condition has length > 1 and only the first element will be used
                                                    26: In if (model_type == "prophet_additive") { ... :
                                                        the condition has length > 1 and only the first element will be used
                                                      27: In if (model_type == "auto_arima") { ... :
                                                          the condition has length > 1 and only the first element will be used
                                                        28: In if (model_type == "prophet_additive") { ... :
                                                            the condition has length > 1 and only the first element will be used
                                                          29: In if (model_type == "prophet_additive") { ... :
                                                              the condition has length > 1 and only the first element will be used
                                                            30: In if (model_type == "prophet_additive") { ... :
                                                                the condition has length > 1 and only the first element will be used
                                                              31: In if (model_type == "auto_arima") { ... :
                                                                  the condition has length > 1 and only the first element will be used
                                                                32: In if (model_type == "prophet_additive") { ... :
                                                                    the condition has length > 1 and only the first element will be used
                                                                  33: In if (model_type == "prophet_additive") { ... :
                                                                      the condition has length > 1 and only the first element will be used
                                                                    34: In if (model_type == "auto_arima") { ... :
                                                                        the condition has length > 1 and only the first element will be used
                                                                      35: In if (model_type == "prophet_additive") { ... :
                                                                          the condition has length > 1 and only the first element will be used
                                                                        36: In if (model_type == "prophet_additive") { ... :
                                                                            the condition has length > 1 and only the first element will be used
                                                                          37: In if (model_type == "prophet_additive") { ... :
                                                                              the condition has length > 1 and only the first element will be used
                                                                            38: In if (model_type == "prophet_additive") { ... :
                                                                                the condition has length > 1 and only the first element will be used
                                                                              39: In if (model_type == "prophet_additive") { ... :
                                                                                  the condition has length > 1 and only the first element will be used
                                                                                40: In if (model_type == "auto_arima") { ... :
                                                                                    the condition has length > 1 and only the first element will be used
                                                                                  41: In if (model_type == "auto_arima") { ... :
                                                                                      the condition has length > 1 and only the first element will be used
                                                                                    42: In if (model_type == "prophet_additive") { ... :
                                                                                        the condition has length > 1 and only the first element will be used
                                                                                      43: In if (model_type == "prophet_additive") { ... :
                                                                                          the condition has length > 1 and only the first element will be used
                                                                                        44: In if (model_type == "prophet_additive") { ... :
                                                                                            the condition has length > 1 and only the first element will be used
                                                                                          45: In if (model_type == "auto_arima") { ... :
                                                                                              the condition has length > 1 and only the first element will be used
                                                                                            46: In if (model_type == "auto_arima") { ... :
                                                                                                the condition has length > 1 and only the first element will be used
                                                                                              47: In if (model_type == "prophet_additive") { ... :
                                                                                                  the condition has length > 1 and only the first element will be used
                                                                                                48: In if (model_type == "prophet_additive") { ... :
                                                                                                    the condition has length > 1 and only the first element will be used
                                                                                                  49: In if (model_type == "prophet_additive") { ... :
                                                                                                      the condition has length > 1 and only the first element will be used
                                                                                                    50: In if (model_type == "auto_arima") { ... :
                                                                                                        the condition has length > 1 and only the first element will be used
                                                                                                      > vis_drake_graph(new_plan_dynamic_branch_test)
                                                                                                      >

Which make me wonder if the workflow is parsing properly the model types:

option_list = list(
  make_option(c("-d", "--directory"), type="character", 
              default=getwd(), 
              help="Directory of the main workflow file", metavar="character"),
  make_option(c("-b", "--outputbucket"), type="character", 
              default="testbucket13245678", 
              help="Bucket to save the forecasts", metavar="character"),
  make_option(c("-c", "--configdir"), type="character", 
              default=paste0(getwd(), "/config/Parms.ini", sep = ""), 
              help="Config file directory", metavar="character"),
  make_option(c("-k", "--keysdirectory"), type="character", 
              default=paste0(getwd(), "/credentials", sep = ""), 
              help="Snowflake credentials directory", metavar="character"),
  make_option(c("-h", "--horizon"), type="integer", 
              default=52, 
              help="Forecasting horizon", metavar="number"),
  make_option(c("-s", "--sessions"), type="integer", 
              default=12, 
              help="Maximum paralell multisessions", metavar="number"),
  make_option(c("-m", "--model_types"), type="character", 
              default=c("auto_arima",
                        "arima_with_yearly_fourier_components",
                        "pophet_additive",
                        "prophet_multiplicative",
                        "arima_with_monthly_fourier_components",
                        "regression_with_arima_errors",
                        "tslm",
                        "theta"),
              help="Model types separated by commas", metavar="number")
  
);

wlandau · 2020-09-14T18:32:38Z

To debug, I suggest running locally with a browser() statement inside your custom accuracy function. Then when you call make(), R will drop you into an interactive debugger and you can interactively access the data and options from inside the function. You could also call debug(accuracy_explorer) or debugonce(accuracy_explorer) right before make(). For more on interactive debugging, I recommend https://rstats.wtf/debugging-r-code.html or https://adv-r.hadley.nz/debugging.html.

edgBR · 2020-09-15T11:02:43Z

Dear @wlandau

Thanks for your suggestion,

I have fix the workflow, it was a typo error in parsing the !!model_types in the training function:

new_plan_dynamic_branch_test <- drake_plan(
  unit_metadata = getMetadata(
    environment = "PROD",
    key_directory = config_parameters$LOCAL_CONFIG$DirectoryKeyCloud_RStudio,
    operex_schema = config_parameters$SF_CONFIG$schema_name, db_src = c(1, 2, 3)
  ),
  distinct_b_numbers = c('xxxxxxxxxxxxx'),
  distinct_db_src = unit_metadata$db_src %>% unique() %>% as.numeric() %>% sort(),
  distinct_snsr_keys = getCountersKeys(
    environment = "PROD",
    key_directory = config_parameters$LOCAL_CONFIG$DirectoryKeyCloud_RStudio,
    operex_schema = config_parameters$SF_CONFIG$schema_name
  ),
  getIndividualData = target(
    getIndividualCounterData(
      environment = "PROD",
      key_directory = config_parameters$RSTUDIO_CLOUD_CONF$KeyDir,
      operex_schema = config_parameters$SF_CONFIG$schema_name,
      package_b_number = distinct_b_numbers,
      counter = distinct_snsr_keys,
      db_src = distinct_db_src, 
      max_forecasting_horizon = argument_parser$horizon
    ),
    dynamic = cross( # Use `dynamic =` instead of `transform =`
      distinct_b_numbers, # no tidy evaluation needed for dynamic branching
      distinct_snsr_keys,
      distinct_db_src
    )
  ),
  processingData = target(
    featureEngineering(
      raw_data = getIndividualData, 
      max_forecasting_horizon = argument_parser$horizon
    ),
    dynamic = map(getIndividualData)
  ),
  training_data = target(
    timeSeriesSplitter(input_data = processingData,
                       max_forecast_horizon = argument_parser$horizon,
                       type = "train"
    ),
    dynamic = map(processingData)
  ),
  testing_data = target(
    timeSeriesSplitter(input_data = processingData,
                       max_forecast_horizon = argument_parser$horizon,
                       type = "test"
    ),
    dynamic = map(processingData)
  ),
  models_training = target( ### split train/validation set
    trainModels(
      input_data = training_data,
      max_forecast_horizon = argument_parser$horizon, 
      model_type = model_type,
      max_multisession_cores = argument_parser$sessions
    ),
    dynamic = map(training_data), # dynamic branching
    transform = map(model_type = !!model_types)
  ),
  accuracy = target(
    accuracy_explorer(
      mode = "test",
      models = models_training,
      max_forecast_horizon = argument_parser$horizon,
      bucket = argument_parser$outputbucket,
      bucket_folder = "/test_lifecounter2",
      testing_data = testing_data
  ), 
  dynamic = map(models_training, testing_data),
  transform = map(models_training)
  ),
  saveModels = target(
    saveModelsS3(
      model = models_training,
      bucket = argument_parser$outputbucket, 
      bucket_folder = "/test_lifecounter2", 
      max_forecasting_horizon = argument_parser$horizon
    ),
    dynamic = map(models_training),
    transform = map(model_type)
  )
  
)

However, now I am getting an error as sometimes I can not estimate the model and then the saving functions and accuracy functions can not be computed:

>subtarget saveModels_regression_with_arima_errors_c635a0a6
[1] "Saving the all-mighty mable"
x fail saveModels_regression_with_arima_errors_259e6ae8
Error: target saveModels_regression_with_arima_errors_259e6ae8 failed.
diagnose(saveModels_regression_with_arima_errors_259e6ae8)error$message:
  object 'models' not found
diagnose(saveModels_regression_with_arima_errors_259e6ae8)error$calls:
  1. └─global::saveModelsS3(...)
2.   ├─base::print(...)
3.   ├─base::paste0(...)
4.   └─base::unique(models$snsr_key)
In addition: Warning messages:
  1: In sqrt(diag(best$var.coef)) : NaNs produced
2: In sqrt(diag(best$var.coef)) : NaNs produced
3: In sqrt(diag(best$var.coef)) : NaNs produced
4: In sqrt(diag(best$var.coef)) : NaNs produced
>

Should I execute a trycatch within the training function, return NULL otherwise and then do not calculate accuracy metrics/save function or there is anyway than drake can handle this by default?

wlandau · 2020-09-15T11:59:31Z

As before, I think this requires more debugging in fable or your custom functions to figure out why the the model cannot run sometimes. A stopgap is keep_going = TRUE in make(), which should allow all the working models and metrics to complete even if some error out.

wlandau · 2020-09-18T11:45:43Z

Closing because the issue is old and no longer specifically drake-related. The comment thread remains open.

edgBR · 2020-09-22T08:12:22Z

Hi @wlandau

Seems that now the workflow was working:

But I get this error in drake:

[1] "Filtering lifecounter for unit xxxxxxxx db_src 1 and counter 105"
[1] "Filtering lifecounter for unit xxxxxxxx  db_src 1 and counter 70"
[32m■[39m finalize getIndividualData
[32m▶[39m dynamic processingData
[32m>[39m subtarget processingData_98530972

 *** caught segfault ***
address 0x100000011, cause 'memory not mapped'

Never seen this before, neither in drake neither running R code. I have google it and it seems related with RCpp but honestly I have no clue.

BR
/Edgar

wlandau · 2020-09-22T12:46:58Z

I'm not sure what could be causing the error. Can you reproduce it with a small/fast example outside drake? I think the next step is to try to recreate the conditions under which this is happening as simply as possible.

edgBR added the type: trouble label Sep 11, 2020

wlandau closed this as completed Sep 18, 2020

edgBR mentioned this issue Oct 1, 2020

Combining tibbles that come from dynamic and transform maps #1327

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Linking multiple mapping multiple dynamic targets to a target #1321

Linking multiple mapping multiple dynamic targets to a target #1321

edgBR commented Sep 11, 2020

wlandau commented Sep 11, 2020 •

edited

Loading

edgBR commented Sep 14, 2020 •

edited

Loading

edgBR commented Sep 14, 2020 •

edited

Loading

wlandau commented Sep 14, 2020

edgBR commented Sep 15, 2020

wlandau commented Sep 15, 2020

wlandau commented Sep 18, 2020

edgBR commented Sep 22, 2020

wlandau commented Sep 22, 2020

Linking multiple mapping multiple dynamic targets to a target #1321

Linking multiple mapping multiple dynamic targets to a target #1321

Comments

edgBR commented Sep 11, 2020

Prework

Description

Reproducible example

Desired result

Session info

wlandau commented Sep 11, 2020 • edited Loading

edgBR commented Sep 14, 2020 • edited Loading

edgBR commented Sep 14, 2020 • edited Loading

wlandau commented Sep 14, 2020

edgBR commented Sep 15, 2020

wlandau commented Sep 15, 2020

wlandau commented Sep 18, 2020

edgBR commented Sep 22, 2020

wlandau commented Sep 22, 2020

wlandau commented Sep 11, 2020 •

edited

Loading

edgBR commented Sep 14, 2020 •

edited

Loading

edgBR commented Sep 14, 2020 •

edited

Loading