Adrian Adrian - 4 months ago 14
R Question

Force mapply to return a list?

Suppose I have a function that creates data frames. I'd like to run that function with different input values, and then rbind the results together into one big data frame, as below:

CreateDataFrame <- function(type="A", n=10, n.true=8) {
data.frame(success=c(rep(TRUE, n.true), rep(FALSE, n - n.true)), type=type)
}
df <- do.call(rbind, lapply(toupper(letters[1:5]), CreateDataFrame))


My CreateDataFrame function takes three arguments. In the example above, the second and third arguments are held constant. I'd like to do the same as above, but have the second and third arguments change on each call. I think I have to use mapply, like this:

mapply("CreateDataFrame", type=toupper(letters[1:5]), n=10, n.true=8:4)


I'm having trouble because mapply isn't returning a list, which prevents me from running
do.call(rbind, mapply(...))
. How can I end up with a single data frame, as I did in the example at the top?

Looks like mapply is returning a matrix of lists. I was expecting it to return a list of data frames. What should I do differently?

Answer

To get a list of data.frames as the return value, set mapply's SIMPLIFY argument to FALSE. (Its default value is TRUE, which directs the function to "attempt to reduce the result to a vector, matrix or higher dimensional array" -- just what you experienced).

res <- mapply("CreateDataFrame", type=toupper(letters[1:5]), n=10, n.true=8:4, 
              SIMPLIFY = FALSE)

identical(class(res), "list")
[1] TRUE