Bryan Bishop
unread,Oct 29, 2009, 4:23:44 PM10/29/09Sign in to reply to author
Sign in to forward
You do not have permission to delete messages in this group
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to ben lipkowitz, kan...@gmail.com, openmanu...@googlegroups.com, sy...@googlegroups.com
On Thu, Oct 29, 2009 at 3:06 PM, ben lipkowitz wrote:
> eventually I want to be able to copy and paste a table from the machinery's
> handbook ebook, then automatically do symbolic regression on the table to
> yield the formula they used in the first place. i dont know if this is
> actually possible, however, and it still seems like a lot of work. oh well.
I completely forgot about that. Just for future reference--
http://code.google.com/p/pygep/wiki/SymbolicRegressionTutorial
Usually formulas from symbolic regression are really, really terrible,
although that's vanilla without anything fixing the terrible
equations. Also, gnuplot has something that is more akin to curve
fitting and is not symbolic regression.
I would be excited to see a sympy symbolic regression library tied up
to a genetic algorithm (or some other search algorithm).
- Bryan
http://heybryan.org/
1 512 203 0507