1% Generated by roxygen2: do not edit by hand 2% Please edit documentation in R/inverse.R 3\name{step_inverse} 4\alias{step_inverse} 5\title{Inverse Transformation} 6\usage{ 7step_inverse( 8 recipe, 9 ..., 10 role = NA, 11 offset = 0, 12 trained = FALSE, 13 columns = NULL, 14 skip = FALSE, 15 id = rand_id("inverse") 16) 17} 18\arguments{ 19\item{recipe}{A recipe object. The step will be added to the 20sequence of operations for this recipe.} 21 22\item{...}{One or more selector functions to choose variables 23for this step. See \code{\link[=selections]{selections()}} for more details.} 24 25\item{role}{Not used by this step since no new variables are 26created.} 27 28\item{offset}{An optional value to add to the data prior to 29logging (to avoid \code{1/0}).} 30 31\item{trained}{A logical to indicate if the quantities for 32preprocessing have been estimated.} 33 34\item{columns}{A character string of variable names that will 35be populated (eventually) by the \code{terms} argument.} 36 37\item{skip}{A logical. Should the step be skipped when the 38recipe is baked by \code{\link[=bake.recipe]{bake.recipe()}}? While all operations are baked 39when \code{\link[=prep.recipe]{prep.recipe()}} is run, some operations may not be able to be 40conducted on new data (e.g. processing the outcome variable(s)). 41Care should be taken when using \code{skip = TRUE} as it may affect 42the computations for subsequent operations.} 43 44\item{id}{A character string that is unique to this step to identify it.} 45} 46\value{ 47An updated version of \code{recipe} with the new step added to the 48sequence of any existing operations. 49} 50\description{ 51\code{step_inverse} creates a \emph{specification} of a recipe 52step that will inverse transform the data. 53} 54\details{ 55When you \code{\link[=tidy]{tidy()}} this step, a tibble with columns \code{terms} 56(the columns that will be affected) is returned. 57} 58\examples{ 59set.seed(313) 60examples <- matrix(runif(40), ncol = 2) 61examples <- data.frame(examples) 62 63rec <- recipe(~ X1 + X2, data = examples) 64 65inverse_trans <- rec \%>\% 66 step_inverse(all_numeric_predictors()) 67 68inverse_obj <- prep(inverse_trans, training = examples) 69 70transformed_te <- bake(inverse_obj, examples) 71plot(examples$X1, transformed_te$X1) 72 73tidy(inverse_trans, number = 1) 74tidy(inverse_obj, number = 1) 75} 76\seealso{ 77Other individual transformation steps: 78\code{\link{step_BoxCox}()}, 79\code{\link{step_YeoJohnson}()}, 80\code{\link{step_bs}()}, 81\code{\link{step_harmonic}()}, 82\code{\link{step_hyperbolic}()}, 83\code{\link{step_invlogit}()}, 84\code{\link{step_logit}()}, 85\code{\link{step_log}()}, 86\code{\link{step_mutate}()}, 87\code{\link{step_ns}()}, 88\code{\link{step_poly}()}, 89\code{\link{step_relu}()}, 90\code{\link{step_sqrt}()} 91} 92\concept{individual transformation steps} 93