R: Sampling a random row from each data frame in a list of data frames

user_18163132 0
Viewed 32

I seem to be having trouble trying to sample a random row from each data frame in my list of data frames.

Here is the code for reproducing the list:

output <- list()

iterations <- 5
for(i in 1:iterations){
  output[[i]] <-  mtcars <- mtcars[sample(nrow(mtcars), size = 15, replace = FALSE), ]

}

After getting the list of data frames, my goal is to take a random row from each of the data frames then add the sampled rows to a numbered list each as data frames.

I think that I may have to use the "map" function but I am not sure how to go about doing this.

Any help at all would be greatly appreciated!

r
2 Answers

You can do this lots of ways, I like the map_dfr function from the purrr package which returns a nice data frame.

library(purrr)

# add a rownumber so we can see we are drawing different rows
mtcars$rownumber <- 1:nrow(mtcars)

# quick list of rows
dflist <- list(mtcars, mtcars, mtcars)

# purrr map solution
map_dfr(dflist, .f = function(df){df[sample(nrow(df),1),]})
#>              mpg cyl  disp  hp drat    wt qsec vs am gear carb rownumber
#> Merc 450SL  17.3   8 275.8 180 3.07 3.730 17.6  0  0    3    3        13
#> Volvo 142E  21.4   4 121.0 109 4.11 2.780 18.6  1  1    4    2        32
#> AMC Javelin 15.2   8 304.0 150 3.15 3.435 17.3  0  0    3    2        23

Created on 2022-02-23 by the reprex package (v2.0.1)

user_49784640

Best to avoid using explicit loops whenever possible. Here's another way using base R:

set.seed(7)
iterations <- 5
sampsize <- 15
nrows <- nrow(mtcars)
samps <- rep(sample(1:nrows, sampsize, replace = FALSE), iterations)
random_rows <- matrix(samps, ncol = 15)
random_rows
     [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12] [,13] [,14] [,15]
[1,]   10   15   24   10   15   24   10   15   24    10    15    24    10    15    24
[2,]   19   26   27   19   26   27   19   26   27    19    26    27    19    26    27
[3,]   28   22   25   28   22   25   28   22   25    28    22    25    28    22    25
[4,]    7    8   12    7    8   12    7    8   12     7     8    12     7     8    12
[5,]    2    3   20    2    3   20    2    3   20     2     3    20     2     3    20
user_35741560
Related