[R] apply() function: margin argument: "2L" versus "2"

[Rui Barradas]

Right, I had missed that one.

To the op: what Michael is saying can be shown with the following example.

x <- array(1:24, c(2,3,4))

r1 <- apply(x, -1, sum)  # remove dim 1
r2 <- apply(x, 2:3, sum) # include dims 2 and 3

all.equal(r1, r2)  # FALSE, different attributes
all(r1 == r2)  # TRUE, equal values

A negative index removes that removes that value from a vector, or 
row/column from a matrix or data.frame, or list element, etc. The index 
must be a number, for instance, data.frame column NAMES can't be used.
c(1, 2, 3)[-2] is c(1, 3).

So, the two forms above should be equivalent, but they're not. The first 
doesn't keep the dim attribute.

Rui Barradas

Em 23-06-2012 00:00, R. Michael Weylandt escreveu:
> On Fri, Jun 22, 2012 at 5:53 PM, Rui Barradas<ruipbarradas at sapo.pt>  wrote:
>> Hello,
>>
>> 2L is an integer, 2 might be or not. In the case of apply(), there is no
>> difference.
>>
>> All possible values for margin? Any possible(*) combination of the
>> dimensions of 'x' in
>>
>> apply(x, margin, function)
>>
>> (*) non-null. If, say, x<- array(0, dim=c(2,3,4)) (3dim) then you can call
>> any of
>>
>> apply(x, 1, fun) or apply(x, 2, fun) or apply(x, 3, fun) but not margin=4;
>> apply(x, 1:2, fun) or apply(x, c(1, 3), fun) or ...
>> apply(x, 1:3, fun)
> You can also get away with negative indexing in apply():
>
> x<- array(1:24, dim = c(2,3,4))
>
> apply(x, -1, sum)
> apply(x, -c(1,3), sum)
>
> but you seem to loose the array structure.
>
> As usual, you can't mix
>
> apply(x, c(-1, 2), sum) # ERROR
>
> It also seems zero doesn't work here:
>
> apply(x, 0, sum) # ERROR
>
> Best,
> Michael
>
>> Hope this helps,
>>
>> Rui Barradas
>>
>> Em 22-06-2012 23:23, Luba G escreveu:
>>
>>> What is the difference of using 2L versus 2 as the margin argument in the
>>> apply() function? Where can I find detailed information on all of the
>>> possible margin arguments?
>>>
>>>> x
>>>       [,1] [,2]
>>> [1,]    1    2
>>> [2,]    3    4
>>> [3,]    5    6
>>> [4,]    7    8
>>> [5,]    9   10
>>>> sqrt(apply(x, *2L*, function(r.st) var(r.st[!is.na(r.st)])))
>>> [1] 3.162278 3.162278
>>>> sqrt(apply(x,* 2*, function(r.st) var(r.st[!is.na(r.st)])))
>>> [1] 3.162278 3.162278
>>>
>>>         [[alternative HTML version deleted]]
>>>
>>> ______________________________________________
>>> R-help at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide
>>> http://www.R-project.org/posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.