Write a function that changes !is.numeric columns in a dataframe =NULL

[Farm]

I need help in writing a function that checks if any column in a data frame isn't numeric and equates that column to a NULL.

[FJCC]

Setting a column to NULL removes it from the data frame. Is your goal simply to remove non-numeric columns or is this a homework assignment for you to practice writing functions?

[Farm]

Yes, my goal is to remove non-numeric columns.
I am practicing the writing of functions,,it's not an assignment

[FJCC]

I would write a function that returns either the column or NULL and then use one of the built in functions like lapply() or map() that iterate over a list. A data frame is just a list with some extra features.

MyFunc <- function(Col) {
  if (is.numeric(Col)) {
    return(Col)
  } else {
    return(NULL)
  }
}
DF <- data.frame(N1 = 1:4, C1 = LETTERS[1:4], 
                 N2 = 10:13, C2 = LETTERS[5:8])
DF
#>   N1 C1 N2 C2
#> 1  1  A 10  E
#> 2  2  B 11  F
#> 3  3  C 12  G
#> 4  4  D 13  H
DF[] <- lapply(DF, MyFunc)
DF
#>   N1 N2
#> 1  1 10
#> 2  2 11
#> 3  3 12
#> 4  4 13

DF <- data.frame(N1 = 1:4, C1 = LETTERS[1:4], 
                 N2 = 10:13, C2 = LETTERS[5:8])
library(purrr)
DF <- map_dfc(DF, MyFunc)
DF
#> # A tibble: 4 x 2
#>      N1    N2
#>   <int> <int>
#> 1     1    10
#> 2     2    11
#> 3     3    12
#> 4     4    13

^{Created on 2022-05-04 by the reprex package (v2.0.1)}

[Farm]

Thanks
kinda helped,
but I would have hoped to get a data frame or matrix because the functions I would use afterward (e.g agnes function) would not work on lists

[FJCC]

The map_dfc function returns a tibble, which is a data frame with some particular behavior when used with tidyverse functions. You can use it as a data frame. If you run the class() function on DF, you will see that it does include data.frame. Any function that does not have methods for tbl_df or tbl will treat DF as a data.frame.

class(DF)
[1] "tbl_df"     "tbl"        "data.frame"

[nirgrahamuk]

an alternative

myfunc <- function(x){
  require(dplyr)
    select_if(x,
              ~!is.numeric(.))
}