I was curious to see if there is a way to craft a for loop to do apply as.numeric() for a certain columns, since the entire DF is considered to be as a character. Perhaps I have a very large data set, and I want to make a certain column, until the very last column, to be applied.

I looked into lapply, but I think I am more confused on that.

DF =

Is there a way to make a for loop where I can do as.numeric() for each column, say starting at column 3 and until the very end? I would be working with large data sets, so I was wondering if there was a way to do that. Cause I dont think I can do as.numeric when there are actual characters in column 1 and 2, and using unlist() makes the problem worse.

Person <- c("First", "Second", "Third", "Fourth")
KLAS <- c("Meyer", "Walmart", "Safeway", "Kroger")
XDXS <- c("33", "34", "35", "22")
ASDD <- c("21", "28", "21", "11")
BBHS <- c("22", "23", "21", "18")
JKJA <- c("12", "19", "11", "8")
DF <- data.frame(Person, KLAS, XDXS, ASDD, BBHS, JKJA)

CodePudding user response：

You can slice only the columns you want and use apply with 2 as the second argument to apply a function along each column:

DF[, 3:ncol(DF)] <- apply(DF[, 3:ncol(DF)], 2, as.numeric)

CodePudding user response：

You could look into this similar question

See below code adapted from one of the answer. You can design a function to check if you can convert the column (otherwise as.numeric() will introduce NAs) and then transform your data:

library(dplyr)

is_all_numeric <- function(x) {
  !any(is.na(suppressWarnings(as.numeric(na.omit(x))))) & is.character(x)
}

DF %>%  mutate_if(is_all_numeric,as.numeric)

  Person    KLAS XDXS ASDD BBHS JKJA
1  First   Meyer   33   21   22   12
2 Second Walmart   34   28   23   19
3  Third Safeway   35   21   21   11
4 Fourth  Kroger   22   11   18    8

'data.frame':   4 obs. of  6 variables:
 $ Person: chr  "First" "Second" "Third" "Fourth"
 $ KLAS  : chr  "Meyer" "Walmart" "Safeway" "Kroger"
 $ XDXS  : num  33 34 35 22
 $ ASDD  : num  21 28 21 11
 $ BBHS  : num  22 23 21 18
 $ JKJA  : num  12 19 11 8