What's the difference between lapply and do.call?

Question

I\'m learning R recently and confused by two function: lapplyand do.call. It seems that they\'re just similar to map function in Lisp.

Answer 1

lapply() is a map-like function. do.call() is different. It is used for passing the arguments to a function in list form instead of having them enumerated. For instance,

> do.call("+",list(4,5))
[1] 9

Answer 2

Although there have been many answers, here is my example for reference. Suppose we have a list of data as :

L=list(c(1,2,3), c(4,5,6))

The function lapply returns a list.

lapply(L, sum) 

The above means something like below.

list( sum( L[[1]]) , sum( L[[2]]))

Now let us do the same thing for do.call

do.call(sum, L) 

It means

sum( L[[1]], L[[2]])

In our example, it returns 21. In short, lapply always returns a list while the return type of do.call really depends on the function executed.

Answer 3

There is a function called Map that may be similar to map in other languages:

• lapply returns a list of the same length as X, each element of which is the result of applying FUN to the corresponding element of X.

• do.call constructs and executes a function call from a name or a function and a list of arguments to be passed to it.

• Map applies a function to the corresponding elements of given vectors... Map is a simple wrapper to mapply which does not attempt to simplify the result, similar to Common Lisp's mapcar (with arguments being recycled, however). Future versions may allow some control of the result type.

---

1. Map is a wrapper around mapply
2. lapply is a special case of mapply
3. Therefore Map and lapply will be similar in many cases.

For example, here is lapply:

lapply(iris, class)
$Sepal.Length
[1] "numeric"

$Sepal.Width
[1] "numeric"

$Petal.Length
[1] "numeric"

$Petal.Width
[1] "numeric"

$Species
[1] "factor"

And the same using Map:

Map(class, iris)
$Sepal.Length
[1] "numeric"

$Sepal.Width
[1] "numeric"

$Petal.Length
[1] "numeric"

$Petal.Width
[1] "numeric"

$Species
[1] "factor"

---

do.call takes a function as input and splatters its other arguments to the function. It is widely used, for example, to assemble lists into simpler structures (often with rbind or cbind).

For example:

x <- lapply(iris, class)
do.call(c, x)
Sepal.Length  Sepal.Width Petal.Length  Petal.Width      Species 
   "numeric"    "numeric"    "numeric"    "numeric"     "factor" 

Answer 4

In most simple words:

1. lapply() applies a given function for each element in a list,so there will be several function calls.

2. do.call() applies a given function to the list as a whole,so there is only one function call.

The best way to learn is to play around with the function examples in the R documentation.

Answer 5

The difference between both are :

lapply(1:n,function,parameters)

=> This send 1,parameters to function => this sends 2,parameters to function and so on

do.call 

Just sends 1…n as a vector and parameters to function

So in apply you have n function calls,in do.call you have just one

Answer 6

lapply applies a function over a list, do.call calls a function with a list of arguments. That looks like quite a difference to me...

To give an example with a list :

X <- list(1:3,4:6,7:9)

With lapply you get the mean of every element in the list like this :

> lapply(X,mean)
[[1]]
[1] 2

[[2]]
[1] 5

[[3]]
[1] 8

do.call gives an error, as mean expects the argument "trim" to be 1.

On the other hand, rbind binds all arguments rowwise. So to bind X rowwise, you do :

> do.call(rbind,X)
     [,1] [,2] [,3]
[1,]    1    2    3
[2,]    4    5    6
[3,]    7    8    9

If you would use lapply, R would apply rbind to every element of the list, giving you this nonsense :

> lapply(X,rbind)
[[1]]
     [,1] [,2] [,3]
[1,]    1    2    3

[[2]]
     [,1] [,2] [,3]
[1,]    4    5    6

[[3]]
     [,1] [,2] [,3]
[1,]    7    8    9

To have something like Map, you need ?mapply, which is something different alltogether. TO get eg the mean of every element in X, but with a different trimming, you could use :

> mapply(mean,X,trim=c(0,0.5,0.1))
[1] 2 5 8