[SciPy-user] question about ODE output and time steps

Rob Clewley rhc28@cornell....
Thu Apr 26 13:10:02 CDT 2007


Hi,

I don't foresee having time to add (probably only some of) PyDSTool to
SciPy or unless someone would graciously give me a hand. Part of that
would involve some very specific feedback on the basic classes we've
implemented and an evaluation of whether they are suitable for general
usage: in particular, our Interval, Pointset, Variable, and Trajectory
classes. To bring on board PyDSTool's ODE solvers in their current
form is also to commit to the use of these basic classes, as there's
not a trivial way to factor out the solvers without losing some of the
fancy functionality (e.g., having autonomous external inputs to a RHS
interpolated from arrays). In my biased opinion, I think that these
basic classes could add value to SciPy in general, but I'm not going
to push it until I get a sense of sufficient support from you guys.
Some technically-minded folks might want to help me improve our
implementations too.

Also, the lsodar routine already wrapped nicely in SloppyCell could
much more easily be incorporated into SciPy with minimal baggage
included.

Anyway, we haven't yet got around our ugly method for generating
C-based right-hand-sides in a platform-independent way, which is to
(mis-)use distutils! We hope to use auto-generated makefiles instead,
and we have some chance of coding that up over the summer. I would
think that would be a deal-breaker for having our integrators in
SciPy, no?

In the meantime, I would encourage anyone who writes for SciPy, has
played with PyDSTool, and is curious about incorporating our solvers
into SciPy, to please continue this discussion with a critical
evaluation of our base classes (recent improvements are in our SVN
repository). You can read more about their design and implementation
at our wiki.

-Rob


More information about the SciPy-user mailing list