Robyn_refresh error: must provide ‘hyperparameters’ in robyn_inputs()

[boxx0r]

Hi,

I am trying to run robyn_refresh on a previous model output (10_105_15), by first recreating the model using fixed hyperparameters, which creates a folder with the solID 1_1_1 json file successfully, including the hyperparameter single values in the json file, indicating the final hyperparameters were successfully taken from the previous model output.

However, once I run robyn_refresh while calling that 1_1_1.json file, it errors out with the following error:

Error in robyn_run(InputCollect = InputCollectRF, iterations = refresh_iters, : Must provide 'hyperparameters' in robyn_inputs()'s output first

Can anybody help?

OutputModels <- robyn_run(

  InputCollect = InputCollect,

  dt_hyper_fixed = dt_hyper_fixed_10_101_15 # using the same hyperparameters from previous model run to recreate

)

Input data has 323 weeks in total: 2018-10-28 to 2024-12-29

 

Initial model is built on rolling window of 304 week: 2018-11-04 to 2024-08-25

 

>>> Calculating response curves for all models' media variables (43)...

 

Creating directory for outputs: /home/xx/ Robyn_202504141952_init/

 

>>> Collecting 1 pareto-optimum results into: /home/xx/Robyn_202504141952_init/

 

>> Exporting general plots into directory...

 

>> Exporting pareto results as CSVs into directory...

 

>>> Exporting one-pagers into directory...

 

Warning message:

“‘timedatectl’ indicates the non-existent timezone name ‘n/a’”

Warning message:

“Your system is mis-configured: ‘/etc/localtime’ is not a symlink”

Warning message:

“‘/etc/localtime’ is not identical to any known timezone file”

Warning message in geom_text(aes(x = max(.data$x), y = 0.5, vjust = -0.5, hjust = 1, :

“All aesthetics have length 1, but the data has 18544 rows.

ℹ Please consider using `annotate()` or provide this layer with data containing

  a single row.”

Exporting charts as: /home/xx/Robyn_202504141952_init/1_1_1.png

 

Warning message in foreach(sid = uniqueSol, .options.RNG = OutputCollect$seed) %dorng% :

“Foreach loop (doParallelMC) had changed the current RNG type: RNG was restored to same type, next state”

>> Exported 1_1_1 as /home/xx/Robyn_202504141952_init/RobynModel-1_1_1.json

 

Successfully recreated model ID: 1_1_1

Robyn <- robyn_refresh(

  json_file = 'Robyn_202504141952_init/RobynModel-1_1_1.json',

  dt_input = model_inputs_raw,

  dt_holidays = dt_prophet_holidays,

  refresh_steps = 1,

  refresh_mode = "auto",

  refresh_iters = 200,

  refresh_trials = 5

)

>>> Recreating 1_1_1

 

Imported JSON file successfully: Robyn_202504141952_init/RobynModel-1_1_1.json

 

>> Running feature engineering...

 

Input data has 323 weeks in total: 2018-10-28 to 2024-12-29

 

Initial model is built on rolling window of 304 week: 2018-11-04 to 2024-08-25

 

>>> Calculating response curves for all models' media variables (43)...

 

Successfully recreated model ID: 1_1_1

 

Input 'refresh_mode' = 'auto' has been deprecated. Changed to 'manual'

 

 

>>> Building refresh model #1 in manual mode

 

>>> New bounds freedom: 0.33%

 

>> Running feature engineering...

 

Error in robyn_run(InputCollect = InputCollectRF, iterations = refresh_iters, : Must provide 'hyperparameters' in robyn_inputs()'s output first

Traceback:

 

1. robyn_run(InputCollect = InputCollectRF, iterations = refresh_iters,

 .     trials = refresh_trials, refresh = TRUE, add_penalty_factor = listOutputPrev[["add_penalty_factor"]],

 .     ts_validation = ts_validation, ...)

2. stop("Must provide 'hyperparameters' in robyn_inputs()'s output first")

3. .handleSimpleError(function (cnd)

 . {

.     watcher$capture_plot_and_output()

.     cnd <- sanitize_call(cnd)

.     watcher$push(cnd)

.     switch(on_error, continue = invokeRestart("eval_continue"),

 .         stop = invokeRestart("eval_stop"), error = NULL)

. }, "Must provide 'hyperparameters' in robyn_inputs()'s output first",

 .     base::quote(robyn_run(InputCollect = InputCollectRF, iterations = refresh_iters,

 .         trials = refresh_trials, refresh = TRUE, add_penalty_factor = listOutputPrev[["add_penalty_factor"]],

 .         ts_validation = ts_validation, ...)))