CODEWITHSUNDEEP
rdplyrtidyr
lost
lost

Reputation: 1669

nest to get list-columns of vectors rather than list-column of data frames

tidyr::nest() creates a list-column of data frames/tibbles:

library(tidyverse)

iris %>% nest(data = c(-Species))
#> # A tibble: 3 × 2
#>   Species    data             
#>   <fct>      <list>           
#> 1 setosa     <tibble [50 × 4]>
#> 2 versicolor <tibble [50 × 4]>
#> 3 virginica  <tibble [50 × 4]>

This is of course useful in lots of cases. In other cases, what I want is to do the same sort of implicit summarizing operation, but instead of creating a single list-column of tibbles, I want to turn each selected column into a list-column of vectors.

What I'm looking for can be accomplished by following nest() with unnest_wider():

iris %>% nest(data = c(-Species)) %>%
  unnest_wider(data)
#> # A tibble: 3 × 5
#>   Species    Sepal.Length Sepal.Width Petal.Length Petal.Width
#>   <fct>       <list<dbl>> <list<dbl>>  <list<dbl>> <list<dbl>>
#> 1 setosa             [50]        [50]         [50]        [50]
#> 2 versicolor         [50]        [50]         [50]        [50]
#> 3 virginica          [50]        [50]         [50]        [50]

It can also be accomplished by a more explicit summarizing operation:

iris %>%
  group_by(Species) %>%
  summarize(across(everything(), vctrs::list_of))
#> # A tibble: 3 × 5
#>   Species    Sepal.Length Sepal.Width Petal.Length Petal.Width
#>   <fct>       <list<dbl>> <list<dbl>>  <list<dbl>> <list<dbl>>
#> 1 setosa             [50]        [50]         [50]        [50]
#> 2 versicolor         [50]        [50]         [50]        [50]
#> 3 virginica          [50]        [50]         [50]        [50]

But both of these seem a little unwieldly, and I'm wondering if there is a more parsimonious or more standard way of doing this in tidyr or related tidyverse packages.

Upvotes: 5

Views: 125

Answers (1)

Nir Graham
Nir Graham

Reputation: 5137

This saves about 30 characters typed :

iris |>
 group_by(Species) |> 
 summarise_all(list)

Upvotes: 0

Related Questions