How can I reuse a fable model after fitting the model especially when used in a *new environment*?

[lbui30]

I have just recently started using fable and fable.prophet when I have run into a very odd situation. As you can see from the code below, the model runs fine when everything is in the global environment but fails to detect or use the exogenous regressors when the code is in a local environment (like a function):

This code works:

ts2 <- readRDS(url("https://github.com/lbui30/reprexdata/blob/main/ts2.RDS?raw=true"))
testmodel <- readRDS(url("https://github.com/lbui30/reprexdata/blob/main/testmodel.RDS?raw=true"))
sampledata <- ts2 %>% select(!any_of(c("ds"))) %>% filter(station_id %in% c("04"))
myholidays <- readRDS(url("https://github.com/lbui30/reprexdata/blob/main/myholidays.RDS?raw=true"))
myforcast <- testmodel %>%
  forecast(h = "4 weeks")
head(myforcast)
#> # A fable: 4 x 6 [7D]
#> # Key:     battalion, station_id, .model [1]
#>   battalion station_id .model   incident_date   callvolume .mean
#>   <chr>     <chr>      <chr>    <date>              <dist> <dbl>
#> 1 5         04         prophetb 2022-11-20    sample[5000]  47.8
#> 2 5         04         prophetb 2022-11-27    sample[5000]  37.1
#> 3 5         04         prophetb 2022-12-04    sample[5000]  47.6
#> 4 5         04         prophetb 2022-12-11    sample[5000]  47.5

But this code doesn't work:

myprepfunction <- function() {
  ts2 <- readRDS(url("https://github.com/lbui30/reprexdata/blob/main/ts2.RDS?raw=true"))
  testmodel <- readRDS(url("https://github.com/lbui30/reprexdata/blob/main/testmodel.RDS?raw=true"))
  sampledata <- ts2 %>% select(!any_of(c("ds"))) %>% filter(station_id %in% c("04"))
  # simplifying for reprex....
  # myholidays <- create_holidays(sampledata)
  myholidays <- readRDS(url("https://github.com/lbui30/reprexdata/blob/main/myholidays.RDS?raw=true"))
  myforcast <- testmodel %>%
    forecast(h = "4 weeks")
  return(myforcast)
}
forecast = myprepfunction()
#> Error in `mutate()`:
#> ! Problem while computing `prophetb = (function (object, ...) ...`.
#> Caused by error in `value[[3L]]()`:
#> ! object 'myholidays' not found
#>   Unable to compute required variables from provided `new_data`.
#>   Does your model require extra variables to produce forecasts?
#> Backtrace:
#>      ▆
#>   1. ├─global myprepfunction()
#>   2. │ └─testmodel %>% forecast(h = "4 weeks")
#>   3. ├─fabletools::forecast(., h = "4 weeks")
#>   4. ├─fabletools:::forecast.mdl_df(., h = "4 weeks")
#>   5. │ └─dplyr::mutate_at(...)
#>   6. │   ├─dplyr::mutate(.tbl, !!!funs)
#>   7. │   └─dplyr:::mutate.data.frame(.tbl, !!!funs)
#>   8. │     └─dplyr:::mutate_cols(.data, dplyr_quosures(...), caller_env = caller_env())
#>   9. │       ├─base::withCallingHandlers(...)
#>  10. │       └─mask$eval_all_mutate(quo)
#>  11. ├─fabletools (local) `<fn>`(...)
#>  12. └─fabletools:::forecast.lst_mdl(...)
#>  13.   └─fabletools:::mapply_maybe_parallel(...)
#>  14.     └─base::mapply(FUN = .f, ..., MoreArgs = MoreArgs, SIMPLIFY = SIMPLIFY)
#>  15.       ├─fabletools (local) `<fn>`(dots[[1L]][[1L]], dots[[2L]][[1L]], h = "4 weeks", point_forecast = `<named list>`)
#>  16.       └─fabletools:::forecast.mdl_ts(...)
#>  17.         └─base::tryCatch(...)
#>  18.           └─base (local) tryCatchList(expr, classes, parentenv, handlers)
#>  19.             ├─base (local) tryCatchOne(...)
#>  20.             │ └─base (local) doTryCatch(return(expr), name, parentenv, handler)
#>  21.             └─base (local) tryCatchList(expr, names[-nh], parentenv, handlers[-nh])
#>  22.               └─base (local) tryCatchOne(expr, names, parentenv, handlers[[1L]])
#>  23.                 └─value[[3L]](cond)
#>  24.                   └─rlang::abort(...)

P.S. While the full context of my problem can be found here, I am new to the r-community so I am not sure if my post has shown up yet: https://community.rstudio.com/t/why-wont-my-prophet-fable-model-with-holidays-forecast-in-a-new-r-environment/153767/1

[mitchelloharawild]

Thanks for reporting this issue - it should work but perhaps there is a problem that needs fixing here. I'll investigate further when I have the time.

[lbui30]

Thanks for the quick reply mitchelloharawild. As soon as the Rstudio post works I will let you know. Also, I suspect the bundle package might be fixed, but I am still in the process of getting it working correctly. https://github.com/rstudio/bundle.

@mitchelloharawild if it does sound like something needs fixing, please let me know as I don't mind reaching out to the prophet package developers if the issue is on their end. :)

Keep you updated,

Lbui