Hi,

I'm interested in both the port of decomposable objects and the
implementation of Boltzmann sampling, in order to generate at random
paths in graphs. I think I could do everything I need from Sage only.

I used decomposable objects from MuPAD-combinat to count and generate
at random paths in graphs.
Since MuPAD is not free, I've implemented my own C++ code that do only
the part of MuPAD-combinat I need: Load a graph in GraphViz format,
build a counting table up to a size n, with MPFR or GMP numbers, and
generate paths according to those numbers. This code is useful for
benchmarks interests as the counting and generation phases are well
separated, which was not the case of the C code produced by the
generateCode method. As I only consider regular languages, It was also
possible to apply some optimizations to reduce the space-complexity.
One of this optimization already depends on Sage for linear algebraic
computations. Hence my interest for the port of decomposable objects
into Sage ;-)

I also try to compare my algorithms to Boltzmann sampling in a current
work with Carine Pivoteau, and I heard it may be integrated to Sage
soon. It will be easier to compare them if there are in the same tool,
so I can help with this too.

Unfortunately, I couldn't be in Marseille for the next Sage Days, but
I'll read this mailing list and will be in the IRC channel.

Thanks for your great work.

Cheers,

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/sage-combinat-devel?hl=en.

Reply via email to