modeltime: create_xreg_recipe – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

create_xreg_recipe

Developer Tools for preparing XREGS (Regressors)

Description

These functions are designed to assist developers in extending the modeltime package. create_xregs_recipe() makes it simple to automate conversion of raw un-encoded features to machine-learning ready features.

Usage

create_xreg_recipe(
  data,
  prepare = TRUE,
  clean_names = TRUE,
  dummy_encode = TRUE,
  one_hot = FALSE
)

Arguments

`data`	A data frame
`prepare`	Whether or not to run `recipes::prep()` on the final recipe. Default is to prepare. User can set this to FALSE to return an un prepared recipe.
`clean_names`	Uses `janitor::clean_names()` to process the names and improve robustness to failure during dummy (one-hot) encoding step.
`dummy_encode`	Should `factors` (categorical data) be
`one_hot`	If `dummy_encode = TRUE`, should the encoding return one column for each feature or one less column than each feature. Default is `FALSE`.

Details

The default recipe contains steps to:

Remove date features
Clean the column names removing spaces and bad characters
Convert ordered factors to regular factors
Convert factors to dummy variables
Remove any variables that have zero variance

Value

A recipe in either prepared or un-prepared format.

Examples

library(dplyr)
library(timetk)
library(recipes)
library(lubridate)

predictors <- m4_monthly %>%
    filter(id == "M750") %>%
    select(-value) %>%
    mutate(month = month(date, label = TRUE))
predictors

# Create default recipe
xreg_recipe_spec <- create_xreg_recipe(predictors, prepare = TRUE)

# Extracts the preprocessed training data from the recipe (used in your fit function)
juice_xreg_recipe(xreg_recipe_spec)

# Applies the prepared recipe to new data (used in your predict function)
bake_xreg_recipe(xreg_recipe_spec, new_data = predictors)

modeltime

The Tidymodels Extension for Time Series Modeling

v0.5.1

MIT + file LICENSE

Authors

Matt Dancho [aut, cre], Business Science [cph]

Initial release