On Jun 23, 5:50 am, Rolf <[email protected]> wrote:
> This is a more general issue to be discussed how to structure
> ProbabilityDistribution class.

+1 to some serious cleanup in this area.  This would be very welcome.

I like option (2), though once you get into it, maybe (3) will have
greater appeal as you actually write some code.

As KDC has pointed out, it would be nice to have a consistent,
Pythonic, easy-to-understand interface to these distributions in Sage,
whether or not we get the actual computations from GSL, NumPy or R, so
I would think hard about a design that would make it easy to use one
of these (or a combination of all three) in the background.

One nit: a student of mine used a distribution backed by GSL to create
random matrices.  It was very hard to doctest, since we could not
specify a seed in predictable way (that we could see).  So if you see
a way to have the mechanisms for "predictable randomness" in the
doctest structure when using these distributions, that would be a big
bonus.

Thanks,
Rob


-- 
To post to this group, send an email to [email protected]
To unsubscribe from this group, send an email to 
[email protected]
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Reply via email to