Hi all,
I'm running a monte carlo test of a neural network tool I've developed,
and it looks like it's going to take a very long time if I run it in R
so I'm interested in translating my code (included below) into something
faster like Fortran (which I'll have to learn from scratch). However,
As nnet is done almost entirely in compiled C, you may well find that
already most of the computation is in a compiled language.
Please look at `Writing R Extensions' and profile your code to find the
bottlenecks.
While you are looking at that manual, please also consider the section on
I have the impression (after only a quick glance at your code, so I may
have missed something) that you are generating multiple datasets and
then using them. As a general strategy this does not work very well. You
will get slightly improved performance with compiled code, but the real
problem