Jason K. Moore
Jason K. Moore
Maybe the only thing to consider here is whether there is a better way to manage the API given you only need to pass in a single argument instead of...
We could do `Problem.__init__(*args, ...)` and then count the number of arguments so that you could skip the gradient function.
We could also make a second `Problem` as `SymProblem` that takes the new arguments. Or do a `Problem.from_symbolic()` alternative initializer.
As a work around until this is implemented, see the hack in #489.
Firstly, this issue has nothing to do with interpolation. If you want to discuss that, then issue #474 is the issue for that information. Secondly, I'm not quite sure what...
1. Yes, you have to explicitly provide inequality bounds. 2. Using `eom_bounds` ensures that IPOPT is used in the expected way for specifying inequalities on the constraints. If setting equality...
Running this locally with minimal other processes eating cpu, you get: ``` moorepants@nandi:opty(move-parallel)$ ipython examples-gallery/beginner/plot_parallel.py Time for non-parallel: 7.072258315001818 Time for parallelized: 1.9007170830009272 ``` So there is a nice speed...
We could disable parallel builds on RTD, but that might significantly impact our chances of hitting the time limit.
I disabled the parallel sphinx gallery build on RTD and get this result: A speed up but not like locally. Maybe the RTD only has two cores.
The build took 15 minutes (which seems the same as prior builds with sphinx gallery parallel): 15 minutes is about at the limit on ReadTheDocs. So I guess we can...