Troubleshooting

Long compilation times

If a solution routine takes surprisingly long to compile but then executes quickly, it may be due to the choice of Taylor-coefficient computation. Some functions in probdiffeq.taylor unroll a (small) loop. To avoid this, use probdiffeq.taylor.autodiff.taylor_mode_scan() (which is implemented with a scan). If the problem persists, reduce the number of derivatives (if that is appropriate for your integration problem) or switch to a different Taylor-coefficient routine. For example, use a Runge-Kutta starter probdiffeq.taylor.estim.make_runge_kutta_starter(). For \(\nu < 5\), switching to Runge-Kutta starters should preserve performance of the solvers. High-order methods, e.g. \(\nu = 9\) seem to rely on taylor_fn=taylor.taylor_mode_fn.

Other problems

Your problem is not discussed here? Please open an issue!