How to remove the grouped rows from the original data frame and put them in new one?-CodePudding

I want to remove the rows which had been grouped from the original data frame and put them into another data frame.

library(dplyr)

Name <- c("Jon", "Jon", "Jon", "Jon", "Jon", "Jon")
school <- c("a", "a", "b", "c", "x", "y")
Age <- c(10, 15, 20, 25, 30, 60)

df <- data.frame(Name, school, Age )

#case_1
dfAvg <- df %>%
group_by(Name, school) %>% 
summarise(across(Age, mean))

I want after this to remove the rows that resulted from group by so my original Df has 4 rows and a new Df has only 1 row that came out from group by.

expected output: 2 data frames: the first one:

jon b 20
jon c 25
jon x 30
jon y 60

and the second one contains the row that resulted after group by:

jon a 12.5

CodePudding user response：

Here is one way that stores your 2 data frames in a list

ibrary(dplyr)
df %>% 
 group_by(Name, school) %>% 
 summarise(n = n() > 1, Age = mean(Age)) %>% 
 split(., .$n)

`summarise()` has grouped output by 'Name'. You can override using the 

$`FALSE`
# A tibble: 4 x 4
# Groups:   Name [1]
  Name  school n       Age
  <chr> <chr>  <lgl> <dbl>
1 Jon   b      FALSE    20
2 Jon   c      FALSE    25
3 Jon   x      FALSE    30
4 Jon   y      FALSE    60

$`TRUE`
# A tibble: 1 x 4
# Groups:   Name [1]
  Name  school n       Age
  <chr> <chr>  <lgl> <dbl>
1 Jon   a      TRUE   12.5

CodePudding user response：

Or you can use the function slice from dplyr package to get the two outputs :

#First data frame :
dfAvg %>% dplyr::slice(-1)
#or
dfAvg %>% dplyr::slice(2:5)

# A tibble: 4 x 3
# Groups:   Name [1]
  Name  school   Age
  <fct> <fct>  <dbl>
1 Jon   b         20
2 Jon   c         25
3 Jon   x         30
4 Jon   y         60

#Second data frame :
dfAvg %>% dplyr::slice(1)

# A tibble: 1 x 3
# Groups:   Name [1]
  Name  school   Age
  <fct> <fct>  <dbl>
1 Jon   a       12.5