fuzzySim (version 4.38)

multConvert: Multiple conversion

Description

This function can simultaneously convert multiple columns of a matrix or data frame, or multiple layers of a SpatRaster.

Usage

multConvert(data, conversion, cols = 1:ncol(data))

Value

The input data with the specified columns converted as specified in 'conversion'.

Arguments

data

A matrix, an object inheriting class data frame, or a SpatRaster containing the columns or layers that need to be converted

conversion

the conversion to apply, e.g. as.factor, scale, log, or a custom-made function

cols

the columns or layers of 'data' to convert

Author

A. Marcia Barbosa

Details

With this function we can change the data type (e.g. convert with as.integer, as.factor or as.character), scale or log-transform several variables in a data frame. By default, all columns in 'data' are converted, but you can specify just some of those in the 'cols' argument. You can also specify your own function to apply, e.g. a division/multiplication of several columns by a given number (see Examples).

Examples

data(rotif.env)

str(rotif.env)


# convert the first 4 columns to character:

converted.rotif.env <- multConvert(data = rotif.env,
conversion = as.character, cols = 1:4)

str(converted.rotif.env)


# divide some columns by 100:

div100 <- function(x) {
  x / 100
}

rotif.env.div100 <- multConvert(data = rotif.env,
conversion = div100, cols = c(6:10, 12:17))

head(rotif.env.div100)


# scale (standardize) continuous variables:

names(rotif.env)

conts <- names(which(sapply(rotif.env[ , 1:17], is.numeric)))

rotif.env.scaled <- multConvert(data = rotif.env,
conversion = scale, cols = conts)

head(rotif.env.scaled)