[R] Incrementing a counter in lapply

Gabor Grothendieck ggrothendieck at gmail.com
Tue Mar 14 18:41:51 CET 2006


On 3/14/06, Thomas Lumley <tlumley at u.washington.edu> wrote:
> On Tue, 14 Mar 2006, Gabor Grothendieck wrote:
>
> > On 3/14/06, Thomas Lumley <tlumley at u.washington.edu> wrote:
> >> On Tue, 14 Mar 2006, Gabor Grothendieck wrote:
> >>
> >>> On 3/14/06, Thomas Lumley <tlumley at u.washington.edu> wrote:
> >>>> It is probably worth pointing out here that the R documentation does not
> >>>> specify the order in which lapply() does the computation.
> >>>>
> >>>
> >>> I suspect that a huge amount of application code takes advantage
> >>> of this order.
> >>>
> >>
> >> I don't.  The order of evaluation is usually not readily observable. You
> >> either have to use <<- to modify an external variable or you have to
> >> produce printed or graphical output where the order matters.
> >>
> >> There's probably some examples, but there are some examples of people
> >> using solve(t(X) %*% W %*% X) %*% W %*% Y to compute regression
> >> coefficients, too.
> >
> > By order do you mean that the result is returned in a random order
> > or that the result is returned in a fixed order but computed it a random
> > order?
> >
>
> It is returned in a fixed order, the first output corresponding to the
> first input.  The documentation does not specify what order it is
> *computed* in.
>
> This in contrast to eapply(), which returns the results in a random order.

In that case I misunderstood.




More information about the R-help mailing list