Jonathan Stickel wrote: > On 9/16/09 [email protected] wrote: >> Date: Tue, 15 Sep 2009 13:14:25 -0700 (PDT) >> From: esteban1 <[email protected]> >> Subject: regdatasmooth >> To: [email protected] >> Message-ID: <[email protected]> >> Content-Type: text/plain; charset=us-ascii >> >> >> Hi, >> >> I have a problem with the regdatasmooth function implemented in the >> data-smoothing-1.2.0 package. Whenever I want to run a demo or the >> function >> itself I get the following error: error: unary operator `'' not >> implemented for `list' operands. >> >> The regdatasmooth function calls the rgdtsmcore function. This function >> works fine, when I run it with my parameters. >> >> But I would like to use regdatasmooth since it optimizes the smoothing >> through lambda and other parameters. >> I allready searched the web and the mailing list but I was not able >> to find >> something. >> >> Thanks for the help! > > > There have been some changes to the optimization functions in Octave > and Octave-forge recently. There is a new fminunc function with a > different implementation in Octave that takes different arguments. To > get the latest regdatasmooth, download it from here: > > http://octave.svn.sourceforge.net/viewvc/octave/trunk/octave-forge/main/data-smoothing/inst/ > > > > Unfortunately, there is a bug in Octave <= 3.2.2 related to cholesky > factorizations that causes fminunc not to work for single variable > functions. A patch is available that may be included in the upcoming > 3.2.3 release (not completely sure). > > As a workaround, you may put the fminunc from /Octave-forge/ in a > place in your path so that it supersedes the fminunc in /Octave/. If > you do that, you should use the previous version of regdatasmooth (the > one that is currently not working for you). > > Let me know if you have more trouble or if you get it working. > > Jonathan > > > P.S. > I am moving this thread to octave-dev since it is about an > octave-forge function. When replying, please remove "help-octave" > from the send list. > > Thanks a lot for the fast reply.
I'm still using the Octave 3.0.1 as a package download from ubuntu. And as far as I understood your mail it doesn't help to upgrade now, since there is a bug in the new fminunc. What I did not understand from your e-mail is which fminunc function I should put into the path. I have only the one in the optim-1.2.0 package (which was already in my path and superseded the other ones) and this one caused the problem. If I download the one in sourceforge.net it complains about an incorrect call to fminunc. As far as I understand the original Octave distribution has no fminunc function ?? But I'm new into this program, so I hope I'm not sure about it. Thanks again for the help Esteban ------------------------------------------------------------------------------ Come build with us! The BlackBerry® Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9-12, 2009. Register now! http://p.sf.net/sfu/devconf _______________________________________________ Octave-dev mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/octave-dev
