Hervé Pagès
2012-Sep-19 20:41 UTC
[Rd] example(smooth.spline) fails when function is NOT byte-compiled
Hi,
example(smooth.spline) fails with the non byte-compiled version of the
smooth.spline function:
> example(smooth.spline)
smth.s> require(graphics)
smth.s> attach(cars)
smth.s> plot(speed, dist, main = "data(cars) & smoothing
splines")
[... more output deleted...]
smth.s> lines(smooth.spline(speed, dist, df=10), lty=2, col =
"red")
Error in smooth.spline(speed, dist, df = 10) :
smoothing parameter value too small
More precisely, here is what happens on a fresh R session (R-2.15.1
configured with --disable-byte-compiled-packages):
** Checking that smooth.spline() is not compiled:
> smooth.spline
[... output deleted...]
<environment: namespace:stats>
** 1st call works:
> res <- smooth.spline(c(4, 4, 7, 7, 8, 9), c(2, 10, 4, 22, 16, 10))
** 2nd call fails:
> res <- smooth.spline(c(4, 4, 7, 7, 8, 9), c(2, 10, 4, 22, 16, 10))
Error in smooth.spline(c(4, 4, 7, 7, 8, 9), c(2, 10, 4, 22, 16, 10)) :
smoothing parameter value too small
** 3rd call (but now with the compiled version of the function) still
fails:
> library(compiler)
> smooth.spline2 <- cmpfun(smooth.spline)
> smooth.spline2
[... output deleted...]
<bytecode: 0x16813a0>
<environment: namespace:stats>
> res2 <- smooth.spline2(c(4, 4, 7, 7, 8, 9), c(2, 10, 4, 22, 16, 10))
Error in smooth.spline2(c(4, 4, 7, 7, 8, 9), c(2, 10, 4, 22, 16, 10)) :
smoothing parameter value too small
Thanks,
H.
My session info:
> sessionInfo()
R version 2.15.1 (2012-06-22)
Platform: x86_64-unknown-linux-gnu (64-bit)
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
[3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
[5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
[7] LC_PAPER=C LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
attached base packages:
[1] compiler stats graphics grDevices utils datasets methods
[8] base
loaded via a namespace (and not attached):
[1] tools_2.15.1
--
Herv? Pag?s
Program in Computational Biology
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N, M1-B514
P.O. Box 19024
Seattle, WA 98109-1024
E-mail: hpages at fhcrc.org
Phone: (206) 667-5791
Fax: (206) 667-1319
Prof Brian Ripley
2012-Sep-20 06:58 UTC
[Rd] example(smooth.spline) fails when function is NOT byte-compiled
As the posting guide asked of you, try R-patched or R-devel before
posting, and check the bug repository.
? smooth.spline() used DUP = FALSE which allowed its compiled C
code to change the function: this was masked by the default
byte-compilation. (PR#14965.)
On 19/09/2012 21:41, Herv? Pag?s wrote:> Hi,
>
> example(smooth.spline) fails with the non byte-compiled version of the
> smooth.spline function:
>
> > example(smooth.spline)
>
> smth.s> require(graphics)
>
> smth.s> attach(cars)
>
> smth.s> plot(speed, dist, main = "data(cars) & smoothing
splines")
>
> [... more output deleted...]
>
> smth.s> lines(smooth.spline(speed, dist, df=10), lty=2, col =
"red")
> Error in smooth.spline(speed, dist, df = 10) :
> smoothing parameter value too small
>
> More precisely, here is what happens on a fresh R session (R-2.15.1
> configured with --disable-byte-compiled-packages):
>
> ** Checking that smooth.spline() is not compiled:
>
> > smooth.spline
> [... output deleted...]
> <environment: namespace:stats>
>
> ** 1st call works:
>
> > res <- smooth.spline(c(4, 4, 7, 7, 8, 9), c(2, 10, 4, 22, 16,
10))
>
>
> ** 2nd call fails:
>
> > res <- smooth.spline(c(4, 4, 7, 7, 8, 9), c(2, 10, 4, 22, 16,
10))
> Error in smooth.spline(c(4, 4, 7, 7, 8, 9), c(2, 10, 4, 22, 16, 10)) :
> smoothing parameter value too small
>
> ** 3rd call (but now with the compiled version of the function) still
> fails:
>
> > library(compiler)
>
> > smooth.spline2 <- cmpfun(smooth.spline)
>
> > smooth.spline2
> [... output deleted...]
> <bytecode: 0x16813a0>
> <environment: namespace:stats>
>
> > res2 <- smooth.spline2(c(4, 4, 7, 7, 8, 9), c(2, 10, 4, 22, 16,
10))
> Error in smooth.spline2(c(4, 4, 7, 7, 8, 9), c(2, 10, 4, 22, 16, 10)) :
> smoothing parameter value too small
>
> Thanks,
> H.
>
> My session info:
>
> > sessionInfo()
> R version 2.15.1 (2012-06-22)
> Platform: x86_64-unknown-linux-gnu (64-bit)
>
> locale:
> [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
> [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
> [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
> [7] LC_PAPER=C LC_NAME=C
> [9] LC_ADDRESS=C LC_TELEPHONE=C
> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
>
> attached base packages:
> [1] compiler stats graphics grDevices utils datasets methods
> [8] base
>
> loaded via a namespace (and not attached):
> [1] tools_2.15.1
>
>
--
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595