Populate occurrence value of variable in R

Question

I have a dataset and need to create a new variable that will populate the row-wise occurrence of value per the variable in the sorted dataset as shown below.

VAR1    VAR2 (to be created)
C1      1
C1      2
C1      3
C2      1
C3      1
C3      2
C4      1
C5      1

Thanks for the help in advance.

Sotos · Accepted Answer · 2016-04-28T22:10:57.553

1

unlist(sapply(rle(as.character(df$VAR1))$lengths,seq))
#[1] 1 2 3 1 1 2 1 1

edited Apr 28 '16 at 22:10

answered Apr 28 '16 at 21:18

Sotos

51,121
6
32
66

I am getting the error "Error in rle(dumdt$Client_ID) : 'x' must be a vector of an atomic type". Can you point out to what I might be doing wrong ? – Vibhor Kalra Apr 28 '16 at 21:59
Apologies. Got it to work. Needed to convert to a character list. Thanks so much. – Vibhor Kalra Apr 28 '16 at 22:09

akrun · Answer 2 · 2016-04-29T03:47:25.730

0

Here is another option using rle and sequence

with(df1, sequence(rle(VAR1)$lengths))
#[1] 1 2 3 1 1 2 1 1

Or with ave

with(df1, ave(seq_along(VAR1), 
        cumsum(c(TRUE, VAR1[-1]!= VAR1[-length(VAR1)])), FUN = seq_along))
#[1] 1 2 3 1 1 2 1 1

Or using rleid from data.table

library(data.table)
setDT(df1)[, VAR2 := seq_len(.N) , by = rleid(VAR1)]

edited Apr 29 '16 at 03:47

answered Apr 29 '16 at 03:34

akrun

874,273
37
540
662

Populate occurrence value of variable in R

2 Answers2