How to subset your dataframe to keep first 3 duplicate rows in R?-CodePudding

I have a table in the following form

enter image description here

I want to keep only the first 3 rows based on duplicate values of V1 & V@ like below:

enter image description here

df %>% distinct(V1,V2) only let me keep the first row.

CodePudding user response：

You may try

library(dplyr)
df %>%
  group_by(V1, V2) %>%
  filter(row_number() <= 3)
     V1    V2    V3
  <dbl> <dbl> <dbl>
1     1     2     4
2     1     2     5
3     1     2     6
4     9     3    10
5     9     3    15
6     9     3    16

CodePudding user response：

With data.table:

library(data.table)
dt <- data.table(c(rep(1, 4), rep(9, 5)), c(rep(2, 4), rep(3, 5)), c(4:7, 10, 15:18)) 
dt[, .SD[1:3], by = c('V1', 'V2')]
   V1 V2 V3
1:  1  2  4
2:  1  2  5
3:  1  2  6
4:  9  3 10
5:  9  3 15
6:  9  3 16

You can use .SD[1:min(3, .N)] if you don't want a row with V3 = NA if there are less than 3 duplicates.