Randomly selecting columns then displaying the rest of the non randomly selected data

Question

The Code is provided below:

df is my data set 128 x 5000

test_random  <- df[, sample(ncol(df), 32)] 
training_random <-  df[-test_random, ]

Error message :

Error in df[test_random, ] :
only 0's may be mixed with negative subscripts

Any thoughts on how to display the training_random subset?

Sal-laS · Accepted Answer · 2018-08-21T18:03:52.277

1

You have almost done it:

test_random  <- df[, -sample(ncol(df), 32)] 


ncol(test_random)

edited Aug 21 '18 at 18:03

answered Aug 21 '18 at 17:27

Sal-laS

If the OP wants both the 32 columns df and the other df, call `sample` just one time: `inx <- sample(etc)`. Then `df[, inx]` and `df[, -inx]` are the two df's the OP wants. – Rui Barradas Aug 21 '18 at 18:41
@Salman yes thank you that worked. thank you so much – Allen Rahrooh Aug 21 '18 at 21:35
@AllenRahrooh could you please acept the answer? – Sal-laS Aug 22 '18 at 03:14
@Salman ok accepted, sorry new to this website – Allen Rahrooh Aug 25 '18 at 23:43

1 Answers1