`epipredict::flusight_hub_formatter()` errors without a `library(epipredict)` call

[dylanhmorris]

epipredict::flusight_hub_formatter() errors if library(epipredict) (or library(epiprocess) or library(epidatasets)) has not been called first. This occurs because it refers to epidatasets::state_census as simply state_census without first importing that name.

Calling library(epipredict) or library(epiprocess) loads epidatasets as a required package, thus masking the issue.

This makes it less straightforward to use epipredict::flusight_hub_formatter() in other R packages (example). I will make a PR with a simple fix.

Reprex

This is simply the flusight_hub_formatter() example from the docs, but without loading epiprocess or epipredict via library(). The internal abbr_to_location() function errors because it cannot find the object name state_census.

library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
weekly_deaths <- epidatasets::covid_case_death_rates %>%
  filter(
    time_value >= as.Date("2021-09-01"),
    geo_value %in% c("ca", "ny", "dc", "ga", "vt")
  ) %>%
  select(geo_value, time_value, death_rate) %>%
  left_join(epidatasets::state_census %>% select(pop, abbr), by = c("geo_value" = "abbr")) %>%
  mutate(deaths = pmax(death_rate / 1e5 * pop * 7, 0)) %>%
  select(-pop, -death_rate) %>%
  group_by(geo_value) %>%
  epiprocess::epi_slide(~ sum(.$deaths), .window_size = 7, .new_col_name = "deaths_7dsum") %>%
  ungroup() %>%
  filter(weekdays(time_value) == "Saturday")
#> Registered S3 method overwritten by 'tsibble':
#>   method               from 
#>   as_tibble.grouped_df dplyr

cdc <- epipredict::cdc_baseline_forecaster(weekly_deaths, "deaths_7dsum")
#> Registered S3 method overwritten by 'epipredict':
#>   method            from   
#>   print.step_naomit recipes
epipredict::flusight_hub_formatter(cdc)
#> Error in `mutate()`:
#> ℹ In argument: `location = abbr_to_location(tolower(geo_value))`.
#> Caused by error in `abbr_to_location()`:
#> ! object 'state_census' not found

^{Created on 2025-09-17 with reprex v2.1.1}