Re: What is the fastest machine for Mathematica?
- To: mathgroup at smc.vnet.net
- Subject: [mg13664] Re: What is the fastest machine for Mathematica?
- From: "Kenneth R. Foster" <kfoster at eniac.seas.upenn.edu>
- Date: Mon, 10 Aug 1998 20:58:33 -0400
- Organization: University of Pennsylvania
- References: <6q3qkj$f19@smc.vnet.net> <6qboem$a1a@smc.vnet.net>
- Sender: owner-wri-mathgroup at wolfram.com
The problem in this case may be associated with the ill-conditioned nature of this problem. Gaussians are not orthogonal functions, and the program may be floundering among several sets of parameter values, all of which provide almost equally good fit. Paul Abbott wrote: > Christian Keysers wrote: > > > The calculations involve fitting of gaussians to data set, and take > > two long at present (2h30min for each data set requiring 16 * 8 fits). > > This seems pretty slow. Instead of a faster machine I would recommend > optimizing the computation. Perhaps you could post a sample dataset > and the sort of fit you are attempting? -- Kenneth R Foster Department of Bioengineering University of Pennsylvania 220 S. 33rd St. Philadelphia PA 19104-6392 215-898-8534 fax 215-573-2071 President IEEE Society on Social Implications of Technology 1997-8 Chair, EMBS Committee on Man and Radiation 1997- ***************************************************************************** SSIT Website: http://www4.ncsu.edu/unity/users/j/jherkert/ COMAR website http://homepage.seas.upenn.edu/~kfoster/comar.htm Blurb for latest book http://homepage.seas.upenn.edu/~kfoster/book.htm Recent Papers http://homepage.seas.upenn.edu/~kfoster/papers.htm