Replicate each row and change one column into binary values

Question

df <- data.frame(n = c(3, 2, 2), 
                 survive = c(2, 1, 2), 
                 a = c(1,1,0), 
                 b = c(0,0,1))

How can I expand the last two columns of the data.frame above, so that each row appears the number of times specified in the column 'n'. And the second column "survive" changes into binary values 0/1 according to the value of "survive"

In other words:

n  survive a  b
3  2       1  0
2  1       1  0
2  2       0  1

To this

survive a  b
1       1  0
1       1  0
0       1  0
1       1  0
0       1  0
1       0  1
1       0  1

Answer 1

Several alternative solutions:

1) Using base R:

rn <- rep(1:nrow(df), df$n)
df2 <- df[rn,]
df2$survive <- as.integer(df2$survive >= ave(rn, rn, FUN = seq_along))

which gives:

> df2[,-1]
   survive a b
1:       1 1 0
2:       1 1 0
3:       0 1 0
4:       1 1 0
5:       0 1 0
6:       1 0 1
7:       1 0 1

---

2) Using the data.table-package:

library(data.table)
df2 <- setDT(df)[, rid := .I
                 ][, .(survive = c(rep(1, survive), rep(0, n - survive)), a, b)
                   , by = rid
                   ][, rid := NULL][]

which gives:

> df2
   survive a b
1:       1 1 0
2:       1 1 0
3:       0 1 0
4:       1 1 0
5:       0 1 0
6:       1 0 1
7:       1 0 1

Or a bit shorter:

df2 <- setDT(df)[, .(survive = c(rep(1, survive), rep(0, n - survive)), a, b), by = 1:nrow(df)
                 ][, nrow := NULL]

---

3) Using the dplyr-package:

library(dplyr)
df %>% 
  mutate(rid = row_number()) %>% 
  .[rep(1:nrow(df), df$n),] %>% 
  group_by(rid) %>% 
  mutate(survive = c(rep(1, unique(survive)), rep(0, unique(n) - unique(survive))) ) %>% 
  ungroup() %>% 
  select(-n, -rid)

which gives:

# A tibble: 7 × 3
  survive     a     b
    <dbl> <dbl> <dbl>
1       1     1     0
2       1     1     0
3       0     1     0
4       1     1     0
5       0     1     0
6       1     0     1
7       1     0     1

---

Used data:

df <- data.frame(n = c(3, 2, 2), 
                 survive = c(2, 1, 2), 
                 a = c(1,1,0), 
                 b = c(0,0,1))

Answer 2

One solution using splitstackshape to expand rows and dplyr,

library(splitstackshape)
library(dplyr)

df %>% 
  mutate(new = 1) %>% 
  expandRows('n') %>% 
  group_by(grp = cumsum(c(1, diff(survive) != 0))) %>% 
  mutate(survive = replace(new, tail(new, n() - survive[1]), 0)) %>% 
  arrange(grp, desc(survive)) %>% 
  ungroup() %>% 
  select(-c(new, grp))

# A tibble: 7 × 3
#  survive     a     b
#    <dbl> <dbl> <dbl>
#1       1     1     0
#2       1     1     0
#3       0     1     0
#4       1     1     0
#5       0     1     0
#6       1     0     1
#7       1     0     1