adjust_epi_recipe: Adjust a step in an 'epi_workflow' or 'epi_recipe'

In cmu-delphi/epipredict: Basic epidemiology forecasting methods

Adjust a step in an epi_workflow or epi_recipe

Description

Make a parameter adjustment to a step in either an epi_workflow or epi_recipe object.

Usage

adjust_epi_recipe(
  x,
  which_step,
  ...,
  blueprint = default_epi_recipe_blueprint()
)

## S3 method for class 'epi_workflow'
adjust_epi_recipe(
  x,
  which_step,
  ...,
  blueprint = default_epi_recipe_blueprint()
)

## S3 method for class 'epi_recipe'
adjust_epi_recipe(
  x,
  which_step,
  ...,
  blueprint = default_epi_recipe_blueprint()
)

Arguments

x	A epi_workflow or epi_recipe object
which_step	the number or name of the step to adjust
...	Used to input a parameter adjustment
blueprint	A hardhat blueprint used for fine tuning the preprocessing.

Details

This function can either adjust a step in a epi_recipe object or a step from a epi_recipe object in an epi_workflow. The step to be adjusted is indicated by either the step number or name (if a name is used, it must be unique). In either case, the argument name and update value must be inputted as .... See the examples below for brief illustrations of the different types of updates.

Value

x, updated with the adjustment to the specified epi_recipe step.

Examples

library(workflows)

jhu <- covid_case_death_rates %>%
  filter(time_value > "2021-11-01", geo_value %in% c("ak", "ca", "ny"))
r <- epi_recipe(jhu) %>%
  step_epi_lag(death_rate, lag = c(0, 7, 14)) %>%
  step_epi_ahead(death_rate, ahead = 7) %>%
  step_epi_naomit()

wf <- epi_workflow(r, parsnip::linear_reg()) %>% fit(jhu)
latest <- jhu %>%
  filter(time_value >= max(time_value) - 14)

# Adjust `step_epi_ahead` to have an ahead value of 14
# in the `epi_workflow`
# Option 1. Using the step number:
wf2 <- wf %>% adjust_epi_recipe(which_step = 2, ahead = 14)
extract_preprocessor(wf2)
# Option 2. Using the step name:
wf3 <- wf %>% adjust_epi_recipe(which_step = "step_epi_ahead", ahead = 14)
extract_preprocessor(wf3)

# Adjust `step_epi_ahead` to have an ahead value of 14
# in the `epi_recipe`
# Option 1. Using the step number
r2 <- r %>% adjust_epi_recipe(which_step = 2, ahead = 14)
r2
# Option 2. Using the step name
r3 <- r %>% adjust_epi_recipe(which_step = "step_epi_ahead", ahead = 14)
r3

cmu-delphi/epipredict documentation built on March 5, 2025, 12:17 p.m.