Split/melt 数据

Question

假设我有以下 data.frame

df <- data.frame(letters = c("a, b", "a", "b", "a, c"), value = c(1, 2, 3, 4))
df
#>   letters value
#> 1    a, b     1
#> 2       a     2
#> 3       b     3
#> 4    a, c     4

我想split/melt到

#>   letters value
#> 1       a     1
#> 2       b     1
#> 3       a     2
#> 4       b     3
#> 5       a     4 
#> 6       c     4

为了每个 letters

有一条记录

Answer 1

你可以试试strsplit。根据 , 拆分字符，然后重复值字段。

char <- strsplit(as.character(df$letters), ',')
data.frame(letter=unlist(char), value=rep(df$value, sapply(char, FUN=length)))


   letter value
#1      a     1
#2      b     1
#3      a     2
#4      b     3
#5      a     4
#6      c     4

根据@docendo discimus 在评论中更新以获得更快的结果，您可以尝试，

char <- strsplit(as.character(df$letters), ',', fixed = T)
data.frame(letter=unlist(char), value=rep(df$value, lengths(char)))

Answer 2

我们可以使用cSplit

 library(splitstackshape)
 cSplit(df, "letters", ", ", "long")
#   letters value
#1:       a     1
#2:       b     1
#3:       a     2
#4:       b     3
#5:       a     4
#6:       c     4

Answer 3

你也可以按照这一步做

df <- data.frame(letters = c("a, b", "a", "b", "a, c"), value = c(1, 2, 3, 4))
l <- do.call(c,strsplit(as.character(df$letters),split = ","))
Indx <- grepl(",",df$letters)
x <- c()
for(i in 1:length(Indx)){
  ifelse(Indx[i],x <- c(x,rep(df$value[i],2)),x<-c(x,df$value[i]))
}

df <- data.frame(l=l,x=x)

Answer 4

library(dplyr)
library(tidyr)

df <- data.frame(letters = c("a, b", "a", "b", "a, c"), value = c(1, 2, 3, 4))

df %>% mutate(letters = strsplit(as.character(letters), ", ")) %>% unnest(letters)

Source: local data frame [6 x 2]

  value letters
  (dbl)   (chr)
1     1       a
2     1       b
3     2       a
4     3       b
5     4       a
6     4       c

Split/melt 数据

Split/melt data

r

plyr

reshape2

dplyr