fuse_one_hot: Take a set of one hot encoded columns (values: 0 or 1) and...

In rtsho/preprocessr: Functions to make your preprocessing life easier.

Description

Take a set of one hot encoded columns (values: 0 or 1) and transform them back into a single factor column.

Usage

fuse_one_hot(vars, out = ".out", missing = "missing", as.integer = FALSE,
  keep = FALSE)

Arguments

vars	Function or formula that returns selected columns from a data.frame. Alternatively, character vector of column names.
out	Name of the new column
missing	Value to use for the rows where all the one hot columns are zero.
as.integer	Whether the output should be integer (or character).
keep	Should the original one hot encoded column be kept

Examples

df <- data.frame(A=c("a1","a1","a2","a3"), B=c(1, 0, 0, 1), C=c(0, 1, 0, 0))
prep <- fuse_one_hot(vars = c('B', 'C'))
prep$fit(df)
prep$transform(df)
# or more succintly
prep$fit_transform(df)

rtsho/preprocessr documentation built on May 29, 2019, 8:58 a.m.