Question

我有以下数据框

group = c("cat", "dog", "horse")
value = c("1", "2", "3")
list = c("siamese,burmese,balinese","corgi,sheltie,collie","arabian,friesian,andalusian" )
df = data.frame(group, value, list)

df
  group value                        list
1   cat     1    siamese,burmese,balinese
2   dog     2        corgi,sheltie,collie
3 horse     3 arabian,friesian,andalusian

并且正在尝试实现这一目标：

  group value       list
1   cat     1    siamese
2   cat     1    burmese
3   cat     1   balinese
4   dog     2      corgi
5   dog     2    sheltie
6   dog     2     collie
7 horse     3    arabian
8 horse     3   friesian
9 horse     3 andalusian

我知道如何汇总一个数据框，但是现在我意识到我不知道如何用逗号分隔的字符串“汇总”。

Answer 1

data.frame(
  group = c("cat", "dog", "horse"),
  value = c("1", "2", "3"),
  list = c("siamese,burmese,balinese","corgi,sheltie,collie","arabian,friesian,andalusian"),
  stringsAsFactors = FALSE
) -> xdf

tidyverse：

tidyr::separate_rows(xdf, list, sep=",")
##   group value       list
## 1   cat     1    siamese
## 2   cat     1    burmese
## 3   cat     1   balinese
## 4   dog     2      corgi
## 5   dog     2    sheltie
## 6   dog     2     collie
## 7 horse     3    arabian
## 8 horse     3   friesian
## 9 horse     3 andalusian

基本R：

do.call(
  rbind.data.frame,
  lapply(1:nrow(xdf), function(idx) {

    data.frame(
      group = xdf[idx, "group"],
      value = xdf[idx, "value"],
      list = strsplit(xdf[idx, "list"], ",")[[1]],
      stringsAsFactors = FALSE
    )

  })
)
##   group value       list
## 1   cat     1    siamese
## 2   cat     1    burmese
## 3   cat     1   balinese
## 4   dog     2      corgi
## 5   dog     2    sheltie
## 6   dog     2     collie
## 7 horse     3    arabian
## 8 horse     3   friesian
## 9 horse     3 andalusian

枪战：

microbenchmark::microbenchmark(

  unnest = transform(xdf, list = strsplit(list, ",")) %>%
    tidyr::unnest(list),

  separate_rows = tidyr::separate_rows(xdf, list, sep=","),

  base = do.call(
    rbind.data.frame,
    lapply(1:nrow(xdf), function(idx) {

      data.frame(
        group = xdf[idx, "group"],
        value = xdf[idx, "value"],
        list = strsplit(xdf[idx, "list"], ",")[[1]],
        stringsAsFactors = FALSE
      )

    })
  )
)
## Unit: microseconds
##           expr      min        lq     mean   median        uq       max neval
##         unnest 3689.890 4280.7045 6326.231 4881.160  6428.508 16670.715   100
##  separate_rows 5093.618 5602.2510 8479.712 6289.193 10352.847 24447.528   100
##           base  872.343  975.1615 1589.915 1099.391  1660.324  6663.132   100

tidyr操作的糟糕表现让我不断感到惊讶。

Answer 2

IIUC我们在R中有unnest

library(dplyr)
library(tidyr)

df = data.frame(group, value, list,stringsAsFactors = F)
df %>%
   transform(list = strsplit(list, ",")) %>%
   unnest(list)
  group value       list
1   cat     1    siamese
2   cat     1    burmese
3   cat     1   balinese
4   dog     2      corgi
5   dog     2    sheltie
6   dog     2     collie
7 horse     3    arabian
8 horse     3   friesian
9 horse     3 andalusian

反向摘要可在数据框中扩展逗号分隔的字符串

2 个答案: