将整个分组的数据框传递到一个函数总结

发布于 2025-02-13 10:08:36 字数 954 浏览 1 评论 0原文

使用新的总结 dplyr 1.0.0可以传递一个函数以汇总比返回tibble，

library(tidyverse)

df <- tibble(
  grp = rep(1:2, each = 5), 
  x = c(rnorm(5, -0.25, 1), rnorm(5, 0, 1.5)),
  y = c(rnorm(5, 0.25, 1), rnorm(5, 0, 0.5)),
)

quibble <- function(x, q = c(0.25, 0.5, 0.75)) {
  tibble(x = quantile(x, q), q = q)
}

df %>% 
  group_by(grp) %>% 
  summarise(quibble(x, c(0.25, 0.5, 0.75)))

我想知道是否可以将功能传递到incoper总结整个分组的数据框架，而不是列的名称。例如

df %>% 
  group_by(grp) %>% 
  summarise(my_function(d))

d是第一个df for grp == 1，然后df for grp ================================= 2。

我认为我知道如何使用purrr :: Map进行操作，

df %>% 
  group_by(grp) %>% 
  nest() %>% 
  mutate(temp = map(data, my_function)) %>% 
  unnest(temp)

但是我想知道是否可以使用总结进行操作。

原文

With the new summarise from dplyr 1.0.0 it is possible to pass a function to summarise than returns a tibble

library(tidyverse)

df <- tibble(
  grp = rep(1:2, each = 5), 
  x = c(rnorm(5, -0.25, 1), rnorm(5, 0, 1.5)),
  y = c(rnorm(5, 0.25, 1), rnorm(5, 0, 0.5)),
)

quibble <- function(x, q = c(0.25, 0.5, 0.75)) {
  tibble(x = quantile(x, q), q = q)
}

df %>% 
  group_by(grp) %>% 
  summarise(quibble(x, c(0.25, 0.5, 0.75)))

I wonder whether it is possible to pass to the function inside summarise the whole grouped data frame instead of the names of the columns.
Something like

df %>% 
  group_by(grp) %>% 
  summarise(my_function(d))

where d is first df for grp == 1 and then df for grp == 2.

I think that I know how to do it with purrr::map

df %>% 
  group_by(grp) %>% 
  nest() %>% 
  mutate(temp = map(data, my_function)) %>% 
  unnest(temp)

but I wonder whether it is possible to do it with summarise.

分享到QQ

分享到微博

如果你对这篇内容有疑问，欢迎到本站社区发帖提问参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。

发布评论

需要登录才能够评论，你可以免费注册一个本站的账号。

要走干脆点 2025-02-20 10:08:36

tibble(gr=c(1,1,2), a=c(1,2,10), b=c(1,2,10)) %>% 
    print() %>% 
# A tibble: 3 x 3
#      gr     a     b
#   <dbl> <dbl> <dbl>
# 1     1     1     1
# 2     1     2     2
# 3     2    10    10
    group_by(gr) %>% 
    summarise(r=sum(cur_data()))
# A tibble: 2 x 2
#      gr     r
#   <dbl> <dbl>
# 1     1     6
# 2     2    20

tibble(gr=c(1,1,2), a=c(1,2,10), b=c(1,2,10)) %>% 
    print() %>% 
# A tibble: 3 x 3
#      gr     a     b
#   <dbl> <dbl> <dbl>
# 1     1     1     1
# 2     1     2     2
# 3     2    10    10
    group_by(gr) %>% 
    summarise(r=sum(cur_data()))
# A tibble: 2 x 2
#      gr     r
#   <dbl> <dbl>
# 1     1     6
# 2     2    20

回复收藏 0 原文

~没有更多了~