On Wed, Apr 23, 2014 at 10:45 PM, Joachim Durchholz <[email protected]> wrote: > Am 23.04.2014 05:52, schrieb Aaron Meurer: > >> One of my worries is not listed here, which is that you are doing >> things completely backwards from good software design with CSymPy, >> which is getting speed first, and something that works later. > > > Now this is an optimization project, so the point is kind of moot. > > However, it's doing optimization backwards. You don't go ahead and optimize > first, you go ahead and benchmark, then identify bottlenecks, then > reimplement that stuff in a faster fashion, then benchmark again to get > feedback on how successful that was. > I'm missing the latter though.
That's exactly how I developed csympy, but I benchmarked it on artificial problems. But I got it to speeds similar to Mathematica and GiNaC (sometimes faster, sometimes slower). Now my goal is to benchmark on real systems and get a little more exposure and repeat the process. Ondrej -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sympy. To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CADDwiVAmY-hQ%2B5sqtvznw4Ww9OnoW3M%3D3LwLY2asuXq5tn9aGg%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
