On Jan 22, 10:56 pm, [EMAIL PROTECTED] wrote: > Arnaud and Terry, > > Great solutions both of you! Much nicer than mine. I particularly like > Arnaud's latest one based on folding because it's so neat and > conceptually simple. For me, it's the closest so far to my goal of the > most elegant solution.
Thanks! It's a great little problem to think of and it helps bring more fun to this list. Sadly work takes over fun during the week, but I will try to improve it at the weekend. > So anyone got an answer to which set of numbers gives the most targets > from 100 onwards say (or from 0 onwards)? Is Python up to the task? I bet it is :) > A thought on that last one. Two ways to improve speed. First of all, > you don't need to rerun from scratch for each target Yes, I've been doing this by writing an 'action' (see my code) that takes note of all reached results. > Secondly, you > can try multiple different sets of numbers at the same time by passing > numpy arrays instead of single values (although you have to give up > the commutativity and division by zero optimisations). Have to think about this. -- Arnaud -- http://mail.python.org/mailman/listinfo/python-list
