CODEWITHSUNDEEP
rdplyrmodeltidyverse
zhiwei li
zhiwei li

Reputation: 1711

Can not use value in a list column in r

I am trying to make many models using list-columns but failed.

df <- tibble(id = 1:5, df = 1:5) %>% 
  nest(data = df)

Only one variable named df in data. I want to extract the value of df from data, and put it to the second argument of ns() function. I tried 3 ways but all failed.

# way 1: use df as second argument in ns()
df %>% 
    mutate(model = map(data, ~ lm(mpg ~ ns(disp, df), data = mtcars)))
# way 2: use .  as second argument in ns()
df %>% 
    mutate(model = map(data, ~ lm(mpg ~ ns(disp, .), data = mtcars))))
# way 3: use .x as second argument in ns()
df %>%  
    mutate(model = map(data, ~ lm(mpg ~ ns(disp, .x), data = mtcars))))

How can I revise the code?

Any help will be highly appreciated!

Upvotes: 0

Views: 34

Answers (1)

Ronak Shah
Ronak Shah

Reputation: 389012

Take df out from . or .x :

library(tidyverse)

df %>% 
   mutate(model = map(data, ~lm(mpg ~ splines::ns(disp, .x$df), data = mtcars)))

#    id   data             model 
#  <int> <list>           <list>
#1     1 <tibble [1 × 1]> <lm>  
#2     2 <tibble [1 × 1]> <lm>  
#3     3 <tibble [1 × 1]> <lm>  
#4     4 <tibble [1 × 1]> <lm>  
#5     5 <tibble [1 × 1]> <lm>

Upvotes: 3

Related Questions