factor_to_df: Convert factor into a data.frame of logicals

In jwutil: Tools for Data Manipulation and Testing

Description

Convert a single factor into a data.frame with multiple true or false fields, one for each factor. The 'vtreat' package may be a better choice for more comprehensive data preparation.

Usage

factor_to_df(fctr, prefix = deparse(substitute(fctr)), sep = "",
  drop_empty = TRUE, na_as_col = TRUE, verbose = FALSE)

factorToDataframeLogical(fctr, prefix = deparse(substitute(fctr)),
  sep = "", drop_empty = TRUE, na_as_col = TRUE, verbose = FALSE)

Arguments

fctr	factor
prefix	defaults to "f" to pre-pend the factor level when constructing the data frame columns names
sep	scalar character, introduced between factor names and levels when forming new data frame column names
drop_empty	logical, if 'TRUE' (the default) factor levels with no associated values are dropped.
na_as_col	logical scalar: if NA data and/or NA levels, then covert to NA strings and expand these as for any other factor
verbose	single logical value, if TRUE then produce verbose messages

Value

data.frame with columns of logicals

Examples

n <- 10
m <- 20
l <- LETTERS[seq_len(n)]
set.seed(1441)
f <- factor(sample(l, m, replace = TRUE), levels = l)
g <- factor_to_df(f, drop_empty = FALSE)
print(g)
stopifnot(nrow(g) == m, ncol(g) == n)
factor_to_df(
  shuffle(factor(shuffle(LETTERS[1:10]))),
  prefix = ""
)
factor_to_df(factor(c(NA, 1, 2, 3)))
factor_to_df(factor(c(NA, 1, 2, 3)), na_as_col = FALSE)

jwutil documentation built on May 7, 2019, 1:03 a.m.