d2m {str2str}

Data-Frame to Matrix

Description

d2m converts a data.frame to a matrix. The user can specify how to convert factors, character vectors, and integer vectors in the data.frame through the internal use of the d2d function. After the call to d2d, d2m simply calls as.matrix.data.frame(rownames.force = TRUE), which will return a matrix of the most complex typeof of any column in the data.frame (most complex to least complex: character, double, integer, logical). Therefore, if any factors or character vectors are left in the data.frame, it will return a character matrix. On the other side of things, if all columns in the data.frame are logical, then it will return a logical matrix. However, if every column in the data.frame is logical except for one factor or character vector, then it will return a character matrix. (If you have a data.frame where 2 columns are the matrix dimnames and one column is the matrix elements, then use d2a()).

Usage

d2m(
  d,
  fct = "chr",
  chr = "chr",
  lgl = "int",
  order.lvl = "alphanum",
  decreasing = FALSE,
  na.lvl = FALSE,
  check = TRUE
)

Arguments

d

data.frame.

fct

character vector of length 1 specifying what factors should be converted to. There are three options: 1) "chr" for converting to character vectors (i.e., factor labels), 2) "int" for converting to integer vectors (i.e., factor codes), or 3) "fct" for keeping the factor as is without any changes.

chr

character vector of length 1 specifying what character vectors should be converted to. There are three options: 1) "fct" for converting to factors (i.e., elements will be factor labels), 2) "int" for converting to integer vectors (i.e., factor codes after first converting to a factor), or 3) "chr" for keeping the character vectors as is without any changes.

lgl

character vector of length 1 specifying what logical vectors should be converted to. There are four options: 1) "fct" for converting to factors (i.e., "TRUE" and "FALSE" will be factor labels), 2) "chr" for converting to character vectors (i.e., elements will be "TRUE" and "FALSE"), 3) "int" for converting to integer vectors (i.e., TRUE = 1; FALSE = 0), and 4) "lgl" for keeping the logical vectors as is without any changes.

order.lvl

character vector of length 1 specifying how you want to order the levels of the factor. The options are "alphanum", which sorts the levels alphanumerically (with NA last); "position", which sorts the levels by the position the level first appears; "frequency", which sorts the levels by their frequency. If any frequencies are tied, then the ties are sorted alphanumerically (with NA last).

decreasing

logical vector of length 1 specifying whether the ordering of the levels should be decreasing (TRUE) rather than increasing (FALSE).

na.lvl

logical vector of length 1 specifying if NA should be considered a level.

check

logical vector of length 1 specifying whether to check the structure of the input arguments. For example, check whether d is a data.frame. This argument is available to allow flexibility in whether the user values informative error messages (TRUE) vs. computational efficiency (FALSE).

Value

matrix with the same dim and dimnames as d. After applying the factor, character vector, and/or integer vector conversions through d2d, the matrix will have typeof = most complex typeof of any column in the modified data.frame.

Examples

x <- d2m(mtcars); str(x)
dat <- as.data.frame(CO2)
x <- d2m(dat); str(x)
x <- d2m(dat, fct = "int"); str(x)