søn, 21 03 2010 kl. 14:24 +1000, skrev Florent Angly: > On 12/03/10 06:51, Søren Hauberg wrote: > > The main problem seems to be that both Octave and 'optim' comes with an > > implementation of > > 'fminbnd'. I don't know which is better, but I don't like the > > duplication. > > > > Ok. I found that the new fminbnd function is part of the development > version of Octave at the moment: > http://hg.savannah.gnu.org/hgweb/octave/file/62bb59f927b1/scripts/optimization/fminbnd.m > > It seems like this implementation of the function uses the Golden search > method combined with parabolic interpolation (like in Matlab's fminbnd). > This means that the new fminbnd function should be faster that the one > in the Octave optim module. > > I just tested the new Octave fminbnd function quickly and its speed and > accuracy seem similar to those obtained with the Matlab function. The > new fminbnd function seems superior to the old one found in optim.
Thanks for looking at this. I have now removed the 'fminbnd' function from the 'optim' package. Søren ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Octave-dev mailing list Octave-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/octave-dev
