Creating serial numbers based on a group unique id in 2 columns

Question

I have this data

cluster<-c(1,1,1,1,2,2,2,2)
structure<-c(1,1,2,3,1,1,1,2)
str<-data.frame(clusters,structures)

i want to create a third column called serial numbers based on the cluster and structure. this is in such a way that i get the following output

serial.number<-c(1,2,1,1,1,2,3,1)
str2<-data.frame(cluster,structure,serial.number)

thankyou

str$serial.number<-c(1,2,1,1,1,2,3,1) will work? – Karthik S Oct 01 '20 at 15:17 — Karthik S, Oct 01 '20 at 15:17

score 2 · Accepted Answer · answered Oct 01 '20 at 23:18

2

Using dplyr

library(dplyr)
df %>%
      group_by(cluster, structure) %>%
      mutate(serial.number = row_number())

or with data.table

library(data.table)
setDT(df)[, serial.number := rowid(cluster, structure)]

answered Oct 01 '20 at 23:18

akrun

score 0 · Answer 2 · answered Oct 01 '20 at 15:16

0

Try ave + seq_along if you are with base R

within(str, serial.number <- ave(structure,cluster,FUN = seq_along))

answered Oct 01 '20 at 15:16

ThomasIsCoding

score 0 · Answer 3 · answered Oct 01 '20 at 15:18

0

I guess you are looking for this:

library(data.table)
df <- setDT(str)
df[, serial.number := seq(.N), by = .(cluster,structure)]

answered Oct 01 '20 at 15:18

peter

works perfectly, but is there an alternative, without using data.table? – Sam Mwenda Oct 01 '20 at 16:05
probably there is an alternative, but I use datatable on purpose. Is there a reason you don't want to use datatable? – peter Oct 01 '20 at 16:14

3 Answers3