[R] How to have NLME return when convergence not reached

[William Dunlap]

You could try using R.utils::withTimeout(expr, timeout=10) to cause an
error when evaluating the expression expr takes longer than 10 seconds.
Wrap that in tryCatch or try to catch the error and examine the outputs of
lapply for ones of class "TimeoutException" to find the ones that took too
long.

E.g.,
  >
  > system.time(z <- lapply(1:10, function(p)tryCatch(withTimeout(timeout=5,
  +     { for(i in seq_len(10^p))log(sqrt(i)) ; p }),
  +     error=function(e)e))
  + )
     user  system elapsed
   13.521   1.516  15.065
  > sapply(z, function(zi)class(zi)[1])
  [1] "integer"          "integer"          "integer"          "integer"
  [5] "integer"          "integer"          "integer"
 "TimeoutException"
  [9] "TimeoutException" "simpleError"
The first 7 iterations finished within 5 seconds, the next 2 timed out,
and the last had a different error (cannot allocate 74.5 Gb vector).

(R.utils's TimeoutException has an exceptionally long printout so I didn't
show the entire result.)



Bill Dunlap
TIBCO Software
wdunlap tibco.com

On Tue, Dec 2, 2014 at 12:57 PM, Ramiro Barrantes <
ramiro at precisionbioassay.com> wrote:

> Hello,
>
> I am trying to fit many hundreds of simulated datasets using NLME (it's
> all in a big loop in R).  Some don't seem to converge.  I am working on
> addressing the issues by perhaps adjusting my simulation, or tweaking
> iteration steps in nlme, etc.  However, when it doesn't converge, NLME just
> hangs, and my program either stalls for hours/days or takes over the
> computer memory and everything crashes eventually.  Is there a way to tell
> nlme to stop when it doesn't seem to be converging somehow?   I have been
> looking at the parameters in nlmeControl() but see nothing obvious.
>
> Thanks in advance,
>
> Ramiro
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> 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.
>

	[[alternative HTML version deleted]]