r - Optimize using multiple consecutive lapply functions? - Stack Overflow

admin管理员组

文章数量:1277257

I am trying to write a function where I apply the same set of functions to a list.

Currently I got this approach to work but it seems very clumsy and I wonder if and how it can be improved:

t1 <- lapply(mylist,
             FUN = myfunc1)

t2 <- lapply(t1,
             FUN = as.data.frame)

t3 <- lapply(t2,
             FUN = t)

t4 <- lapply(t3,
             FUN = myfunc2)

So I first need to extract objects from a list, turn them into data frames, then transpose them and then finally apply the real function.

I am trying to write a function where I apply the same set of functions to a list.

Currently I got this approach to work but it seems very clumsy and I wonder if and how it can be improved:

t1 <- lapply(mylist,
             FUN = myfunc1)

t2 <- lapply(t1,
             FUN = as.data.frame)

t3 <- lapply(t2,
             FUN = t)

t4 <- lapply(t3,
             FUN = myfunc2)

So I first need to extract objects from a list, turn them into data frames, then transpose them and then finally apply the real function.

• r
• lapply

Share Improve this question Follow asked Feb 24 at 12:40 keep_swimmingkeep_swimming 3311 silver badge66 bronze badges 4

• 3 You don't need to call lapply() repeatedly, just create an assigned (or anonymous) function of your processing steps and do it in a single call. You may be able to get the data into an appropriate format without needing to transpose a data frame but you'd need to provide a reproducible example for better advice. – lroha Commented Feb 24 at 12:49
• 2 lapply(lst, \(x) as.data.frame(x) |> t() |> sqrt()), for instance. – jay.sf Commented Feb 24 at 13:13
• Why turn the data into a df, then transpose? The transpose of a df is not a df, if you have a matrix keep it a matrix because after t() that's what you are going to get. – Rui Barradas Commented Feb 24 at 13:47
• What is mylist? A list of data frames? Why not binding them to a single matrix, data.frame? – Friede Commented Feb 24 at 14:46

Add a comment  | 

2 Answers 2

Sorted by: Reset to default 1

As mentioned in the comments, you can use one instance of lapply to do all of this, you just need to use a custom function in the FUN argument. You can do so with function or \:

lapply(mylist, FUN = function(x) myfunc2(t(as.data.frame(myfunc1(x)))))

Because nested functions can be a bit hard to read, you can use a pipe to decompose your sequence of functions without changing the order of compilation:

lapply(mylist, FUN = \(x){
  x |>
    myfunc1() |>
    as.data.frame() |>
    t() |>
    myfunc2()
})

1) We can use Reduce on a list of functions:

L <- as.list(1:3)
Reduce(lapply, list(sqrt, exp), init = L)

2) Also

lapply(L, \(x) exp(sqrt(x)))

3) or using magrittr

library(magrittr)

lapply(L, . %>% sqrt %>% exp)

4) with purrr. In this case reverse the order of the functions.

library(purrr)

map(L,  compose(exp, sqrt))

5) Using Compose from functional is similar to (4) except for the order of functions.

library(functional)

lapply(L, Compose(sqrt, exp))

本文标签： rOptimize using multiple consecutive lapply functionsStack Overflow