Subject: denominator of really complicated expression
From: Ryan Krauss
Date: Wed, 23 Nov 2005 08:32:51 -0500
I use the SciPy Python module for all my numerical analysis and Python
as my general programing language. For this particular problem, I
plan to use SciPy to do some numerical optimization.
Basically, Python is the language I am most comfortable with and I am
trying to use Maxima only to do the symbolic part of this work.
I have done all that things I want to do for a farily simple example
and I am just running into these problems with a more realistic
problem.
If I was going to try and do everything in Maxima, the first thing I
would need to know if how to do numerical root finding and
optimization using things like Newton-Raphson and Nelder-Mead.
Ryan
On 11/23/05, Richard Fateman wrote:
> It is quite reasonable to expect Python or some other
> language processor to just break when given such a
> large expression to parse.
>
> If what you want to do is evaluate this expression
> for some particular values of the variables, I suggest
> you just do it in maxima. You can even convert it to
> lisp and compile to machine language.
>
> The chance that an expression so large suffers from
> disasterous numerical instability is non-zero. It may
> be appropriate to rearrange using horners rule, too.
>
> What do you plan to use python for?
> RJF
>
>
> Ryan Krauss wrote:
>
> > So, following Richard's advice, I am able to get my calculations to
> > run without crashing and much faster than before. I have one
> > additional problem though. In the end, I need to output these
> > expressions to a file so that they can be read into a Python program.
> > Running stringout(filename,expression) takes 5-15 minutes for each of
> > my expressions (there are currently 4 of them and it takes almost an
> ...
>