Hi Maarten, I appreciate your example. I am doing modifications to accommodate the fast_callable. The only thing is that I have to provide all the symbolic variables, and I have several of them....
The jacobian is a square matrix and I will try it via a list of lists of fast_callable functions. > The reason why you think fortran is easier is probably because you know > fortran already but, don't know cython yet (wild guess about your > background). I actually tried Cython before, on the very same problem (but I don't claim that I know it). I worked for a static example. I failed to figure out a way to pass symbolic stuff down to Cython on which the symbolic system is re-sized as the degree of approximation is raised. In may mind it was easy to stream-out the equations to a file, turn it into Fortran, compile and call a piece of symbolic expression (matrix or vector) as a function. Lazy and fast. If you print([expression]) it is almost Fortran. Maybe a print_fortran([expression]) is what I am looking for. > Hope this helps, Sure it helped! Thanks. Regards, Guilherme -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
