[R] multinom(nnet) analogy for biglm package?

[Prof Brian Ripley]

Here is the direct way:

> .Machine$sizeof.pointer
[1] 8

on a 64-bit system.  You can also figure it out from the size of the 
Ncells, clearly 28 bytes in your example.

You seem to believe a multinomial logistic regression is a GLM: it is not.


On Thu, 21 Dec 2006, Yeh, Richard C wrote:

> OK, well, seeing Thomas Lumley's post earlier today, I figured out the
> answer to #4:
>
>> gc()
>           used  (Mb) gc trigger  (Mb)  max used  (Mb)
> Ncells  1115191  29.8    3469679  92.7  13981968 373.4
> Vcells 14796791 112.9   79783730 608.8 124640525 951.0
>> c <- rnorm(1e9)
> Error in rnorm(1e+09) : cannot allocate vector of length 1000000000
>
> I am using:
>
> R version 2.4.0 Patched (2006-10-03 r39576)
>
>
> 212-933-3305 / richard.c.yeh at bankofamerica.com
> -----Original Message-----
> suspect both copies of R were compiled as 32-bit (but I don't know how
> to verify this).
>

-- 
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