This should be simple, but I can't work it out ...
I want to apply a function, eg. mean to each of the lower lists, eg. I want to return
$
1$
1$a [1] value_of_mean 1$
1$b [1] value_of_mean 2$
2$
2$a [1] value_of_mean 3$
2$b [1] value_of_mean 4
I am trying to use the purrr package
require(purrr)
mylist <- list("1"=list(a=runif(10), b=runif(10)), "2"=list(a=runif(10), b=runif(10)))
map(mylist, mean)
You can call map twice to accomplish this. This will work through each list in the top-level list and perform the mean on each element of that top-level list, i.e., the bottom-level lists:
mylist %>% map(~ map(.x, mean))
that gives you this:
$`1`
$`1`$a
[1] 0.5734347
$`1`$b
[1] 0.5321065
$`2`
$`2`$a
[1] 0.483521
$`2`$b
[1] 0.5138651
You don't have to use map twice, as modify_depth was introduced to tidy-up the double map call.
library(purrr)
set.seed(123)
mylist <- list("1"=list(a=runif(10), b=runif(10)), "2"=list(a=runif(10), b=runif(10)))
modify_depth(mylist, 2, mean)
resulting in:
> modify_depth(mylist, 2, mean)
$`1`
$`1`$a
[1] 0.5782475
$`1`$b
[1] 0.5233693
$`2`
$`2`$a
[1] 0.6155837
$`2`$b
[1] 0.537858
This is the same as
mylist %>% map(~ map(.x, mean))
resulting in:
> mylist %>% map(~ map(.x, mean))
$`1`
$`1`$a
[1] 0.5782475
$`1`$b
[1] 0.5233693
$`2`
$`2`$a
[1] 0.6155837
$`2`$b
[1] 0.537858
