Writing dplyr function to pass to group_map

Question

The problem in question would be to apply the function f to each group of a tibble. It is a simpler way to do this, but I would like to solve the problem using the group_map() function.

Data used: starwars of the dplyr package.

What I want is to get an average of the height variable for a grouped tibble considering the variables gender and species. I know the problem could be easily solved by doing:

starwars %>% group_by(gender, species) %>% summarise(mean = mean(height, na.rm = TRUE))

However, my desire is to implement summarise(mean = mean(height, na.rm = TRUE)) in a function and send to group_map().

I tried to create the f() function that gets the data argument which is a tibble object with the previously defined groups. The second argument of the f() function would be ... so that I could pass the variables of interest from data to f().

f <- function(dados, ...){ dados %>% summarise(mean = mean(..., na.rm = TRUE)) } starwars %>% group_by(gender, species) %>% group_map(.tbl = ., .f = ~f(dados = .x), height)

You need to use tidyeval in your function. See the vignette on it — camille
– camille, Commented May 16, 2019 at 13:38
Using ... in your example implies that you want to get means of multiple columns. Is that the case, or do you want to just pass in one column to get the mean of? — camille
– camille, Commented May 16, 2019 at 13:47
Hi Camile, thanks for the info. tidyeval serves me but I'll think how to do for multiple variables. It can be a good exercise. — Pedro Rafael
– Pedro Rafael, Commented May 16, 2019 at 13:58
See Chapter 8 of the Tidy Evaluation book. It is very close to what you are trying to do. — Artem Sokolov
– Artem Sokolov, Commented May 16, 2019 at 14:22

Pedro Rafael · Accepted Answer · 2019-05-16 16:43:03Z

Solutions:

func_1 <- function(dados, var, ...){ var_interesse <- enquo(var) dots <- enquos(...) # Could be attributed direct reference ... dados %>% group_by(!!!dots) %>% summarise(media = mean(x = !!var_interesse, na.rm = TRUE)) } starwars %>% func_1(var = height, gender, species)

or

func_2 <- function(dados, var){ var_interesse <- enquo(var) #dots <- enquos(...) dados %>% summarise(media = mean(x = !!var_interesse, na.rm = TRUE)) } agrupamento <- starwars %>% group_by(gender, species) agrupamento %>% group_map(.tbl = ., .f = ~func_2(dados = .x, var = height))

Collectives™ on Stack Overflow

Writing dplyr function to pass to group_map

1 Answer 1

Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Related