Error: could not find function "parse_model_rhs"

[mathias-314]

Brief description of the problem

Hi,

thank you for your great work in supporting and optimizing in time series analytics. I would like to study your package fasster with a code snippet from README.md and get the above mentioned error, when I run the following reprex.

I would be very happy, if you can help me to find the solution for this
regards
Mathias

---

library(fasster)
library(tidyverse)
library(lubridate)
library(tsibble)
library(fable)
library(fabletools)

elec_tr <- tsibbledata::vic_elec %>%
filter(
Time < lubridate::ymd("2012-03-01")
) %>%
mutate(WorkDay = wday(Time) %in% 2:6 & !Holiday)

elec_fit <- elec_tr %>%
model(
fasster = FASSTER(log(Demand) ~
WorkDay %S% (fourier(48, 16) + trend(1)) + Temperature + I(Temperature^2)
)

[stephstammel]

This is super late, but who knows who will see it one day?

I believe that the issues is that fable tools:: made a change that stopped exporting some functions. FASSTER relied on those so that's why it broke.

As a get-around I downloaded the following two files from fable tools/R and sourced() them at the beginning of my analysis.

It's a duct tape / hack solution, but it's worked for me!

https://github.com/tidyverts/fabletools/blob/main/R/parse.R
https://github.com/tidyverts/fabletools/blob/main/R/traverse.R

@mitchelloharawild would you like me to make. pull request for this?

[mitchelloharawild]

Thanks Steph, it's been a while since I've looked into this package (6 years it seems!) - mostly since the {dlm} package was archived of CRAN which {fasster} heavily depends on.

That said, the model should still be competitive with many statistical forecasting methods and it would be nice to have the package working again (especially since it seems {dlm} is back on CRAN).

Are you able to fix it without duct tape and make a PR?
If not I'll steal some time from other projects and fix this up.