Hi Dhruv! On Sat, Apr 18, 2009 at 1:10 PM, dhruv <lasnabvu...@gmail.com> wrote: > > Hello all, > > I want to build a class to describe a generating function (as used in > the context of discrete probability distributions). It would have > exactly the functionality of the Poly class, have a normalization > constraint on the coefficients, have a few more special methods that > I'd define, and be created with a different set of arguments. I want > to be able to write > >>>> gf = GeneratingFunction(10,4,4,2) # 0.5 + 0.2 x + 0.2 x^2 + 0.1 x^3 >>>> print sum(gf.coeffs) # normalized coefficients > 1.0 >>>> print gf.expected_value() # a method i've written > 0.9 >>>> two_moment = gf * gf # usual arithmetic for Poly instances >>>> print sum(two_moment.coeffs) > 1.0 >>>> print two_moment.expected_value() > 1.8 > > My first thought was to subclass the Poly class, force the > coefficients to be normalized, pass appropriate arguments to > Poly.__init__, and write methods like GeneratingFunction.expected_value > () for the added functionality I want. > > But there seems to be no Poly.__init__ method for me to override. If > I override Poly.__new__ to accept the syntax I want to use (where the > kth argument is the count of the kth item in the collection -- perfect > for probability problems) then arithmetic for my generating functions > breaks, as the __add__, __mul__, &c. methods call a "class method" > which expect the arguments usually passed to Poly.__new__ and none > other.
I don't understand the problem here exactly (e.g. why __add__ stops working), as here is what I would do: create GeneratingFunction.__new__() with your own arguments and then call Poly.__new__() from it, with any preprocessing that you need. See for example Piecewise.__new__() how this could be done. > > Can I get the functionality I want by forcing > GeneratingFunction.__new__ to perform all the same checks and magic on > its argument that Poly.__new__ is already doing? Is there an easier > way? I think the above should be easy. Let us know if it works. Ondrej --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "sympy" group. To post to this group, send email to sympy@googlegroups.com To unsubscribe from this group, send email to sympy+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sympy?hl=en -~----------~----~----~----~------~----~------~--~---
