[R] Change Function based on ifelse() condtion

Christofer Bogaso bogaso.christofer at gmail.com
Sun Mar 4 19:18:21 CET 2018


Thanks Eric, this is working.

On Sun, Mar 4, 2018 at 11:28 PM, Eric Berger <ericjberger at gmail.com> wrote:
> The reason that it works for Apply_MC=TRUE is that in that case you call
> mclapply(X,FUN,...) and
> the mclapply() function strips off the mc.cores argument from the "..." list
> before calling FUN, so FUN is being called with zero arguments, exactly as
> it is declared.
>
> A quick workaround is to change the line
>
> Lapply_me(as.list(1:4), function(xx) {
>
> to
>
> Lapply_me(as.list(1:4), function(xx,dummyList) {
>
> HTH,
> Eric
>
>
> On Sun, Mar 4, 2018 at 7:21 PM, Christofer Bogaso
> <bogaso.christofer at gmail.com> wrote:
>>
>> Below is my full implementation (tried to make it simple as for
>> demonstration)
>>
>> Lapply_me = function(X = X, FUN = FUN, Apply_MC = FALSE, ...) {
>> if (Apply_MC) {
>> return(mclapply(X, FUN, ...))
>> } else {
>> if (any(names(list(...)) == 'mc.cores')) {
>> myList = list(...)[!names(list(...)) %in% 'mc.cores']
>> }
>> return(lapply(X, FUN, myList))
>> }
>> }
>>
>>
>> Lapply_me(as.list(1:4), function(xx) {
>> if (xx == 1) return('a')
>> if (xx == 2) return('b')
>> if (xx == 3) return('c')
>> if (xx == 4) return('d')
>> }, Apply_MC = FALSE, mc.cores = 2)
>>
>> Error message :
>>
>> Error in FUN(X[[i]], ...) : unused argument (list())
>>
>> Kindly note that, with Apply_MC = TRUE, it is working perfectly.
>>
>> On Sun, Mar 4, 2018 at 10:45 PM, Eric Berger <ericjberger at gmail.com>
>> wrote:
>> > That's fine. The issue is how you called Lapply_me(). What did you pass
>> > as
>> > the argument to FUN?
>> > And if you did not pass anything that how is FUN declared?
>> > You have not shown that in your email.
>> >
>> >
>> >
>> >
>> > On Sun, Mar 4, 2018 at 7:11 PM, Christofer Bogaso
>> > <bogaso.christofer at gmail.com> wrote:
>> >>
>> >> My modified function looks below :
>> >>
>> >> Lapply_me = function(X = X, FUN = FUN, Apply_MC = FALSE, ...) {
>> >> if (Apply_MC) {
>> >> return(mclapply(X, FUN, ...))
>> >> } else {
>> >> if (any(names(list(...)) == 'mc.cores')) {
>> >> myList = list(...)[!names(list(...)) %in% 'mc.cores']
>> >> }
>> >> return(lapply(X, FUN, myList))
>> >> }
>> >> }
>> >>
>> >> Here, I am not passing ... anymore rather passing myList
>> >>
>> >> On Sun, Mar 4, 2018 at 10:37 PM, Eric Berger <ericjberger at gmail.com>
>> >> wrote:
>> >> > Hi Christofer,
>> >> > Before you made the change that I suggested, your program was
>> >> > stopping
>> >> > at
>> >> > the statement: list(...) = list(..) .etc
>> >> > This means that it never tried to execute the statement:
>> >> > return(lapply(X,FUN,...))
>> >> > Now that you have made the change, it gets past the first statement
>> >> > and
>> >> > tries to execute the statement: return(lapply(X,FUN,...)).
>> >> > That attempt is generating the error message because whatever you are
>> >> > passing in as the FUN argument is not expecting extra arguments.
>> >> >
>> >> > HTH,
>> >> > Eric
>> >> >
>> >> >
>> >> > On Sun, Mar 4, 2018 at 6:52 PM, Christofer Bogaso
>> >> > <bogaso.christofer at gmail.com> wrote:
>> >> >>
>> >> >> @Eric - with this approach I am getting below error :
>> >> >>
>> >> >> Error in FUN(X[[i]], ...) : unused argument (list())
>> >> >>
>> >> >> On Sun, Mar 4, 2018 at 10:18 PM, Eric Berger <ericjberger at gmail.com>
>> >> >> wrote:
>> >> >> > Hi Christofer,
>> >> >> > You cannot assign to list(...). You can do the following
>> >> >> >
>> >> >> > myList <- list(...)[!names(list(...)) %in% 'mc.cores']
>> >> >> >
>> >> >> > HTH,
>> >> >> > Eric
>> >> >> >
>> >> >> > On Sun, Mar 4, 2018 at 6:38 PM, Christofer Bogaso
>> >> >> > <bogaso.christofer at gmail.com> wrote:
>> >> >> >>
>> >> >> >> Hi,
>> >> >> >>
>> >> >> >> As an example, I want to create below kind of custom Function
>> >> >> >> which
>> >> >> >> either be mclapply pr lapply
>> >> >> >>
>> >> >> >> Lapply_me = function(X = X, FUN = FUN, ..., Apply_MC = FALSE) {
>> >> >> >> if (Apply_MC) {
>> >> >> >> return(mclapply(X, FUN, ...))
>> >> >> >> } else {
>> >> >> >> if (any(names(list(...)) == 'mc.cores')) {
>> >> >> >> list(...) = list(...)[!names(list(...)) %in% 'mc.cores']
>> >> >> >> }
>> >> >> >> return(lapply(X, FUN, ...))
>> >> >> >> }
>> >> >> >> }
>> >> >> >>
>> >> >> >> However when Apply_MC = FALSE it generates below error saying :
>> >> >> >>
>> >> >> >>   '...' used in an incorrect context
>> >> >> >>
>> >> >> >>
>> >> >> >> Appreciate if you can help me with the correct approach. Thanks,
>> >> >> >>
>> >> >> >>
>> >> >> >> On Sun, Mar 4, 2018 at 9:34 PM, Duncan Murdoch
>> >> >> >> <murdoch.duncan at gmail.com>
>> >> >> >> wrote:
>> >> >> >> > On 04/03/2018 10:39 AM, Christofer Bogaso wrote:
>> >> >> >> >>
>> >> >> >> >> Hi again,
>> >> >> >> >>
>> >> >> >> >> I am looking for some way to alternately use 2 related
>> >> >> >> >> functions,
>> >> >> >> >> based on some ifelse() condition.
>> >> >> >> >>
>> >> >> >> >> For example, I have 2 functions mclapply() and lapply()
>> >> >> >> >>
>> >> >> >> >> However, mclapply() function has one extra parameter
>> >> >> >> >> 'mc.cores'
>> >> >> >> >> which
>> >> >> >> >> lapply doesnt not have.
>> >> >> >> >>
>> >> >> >> >> I know when mc.cores = 1, these 2 functions are essentially
>> >> >> >> >> same,
>> >> >> >> >> however I am looking for more general way to control them
>> >> >> >> >> within
>> >> >> >> >> ifelse() constion
>> >> >> >> >>
>> >> >> >> >> Can someone please help me how can I use them within ifelse()
>> >> >> >> >> condition.
>> >> >> >> >
>> >> >> >> >
>> >> >> >> > Don't.  ifelse() usually evaluates *both* the true and false
>> >> >> >> > values,
>> >> >> >> > and
>> >> >> >> > then selects entries from each.  Just use an if statement.
>> >> >> >> >
>> >> >> >> > Duncan Murdoch
>> >> >> >>
>> >> >> >> ______________________________________________
>> >> >> >> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
>> >> >> >> 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.
>> >> >> >
>> >> >> >
>> >> >
>> >> >
>> >
>> >
>
>



More information about the R-help mailing list