Oscar Dowson
Oscar Dowson
Thinking on this, I should probably merge this first into MathOptSymbolicAD.jl, get it working, and then we can add MathOptSymbolicAD as `MOI.Nonlinear.SymbolicAD`.
Closing in favor of MathOptSymboliAD for now.
I won't dissuade you from tinkering with the code to try and add this, and I will happily review a PR if you make one :smile: With that being said......
> I don't know what would be easier: adding a DCP side to JuMP (i.e., bringing some of Convex into JuMP) or writing a smaller JuMP-like package just for DCP...
> Seems like SciML is working on this already: https://github.com/SciML/SymbolicAnalysis.jl That's a long way from being a usable replacement to Convex.jl. There is definitely room to explore multiple approaches, so...
> As there seems to be no bridge > It would be great if Cumulative (and Path) could also get a MOI bridge Correct, see: > I haven't done Path...
This looks useful: https://www.diva-portal.org/smash/get/diva2:1041663/FULLTEXT01.pdf
> the time it takes for the MILP solver because the MILP reformulation is not appropriate ? I assume this. The cumulative constraint is non-trivial to reformulate. (See PDF above.)
I think I'm going to close this issue as out-of-scope because the reformulation is very non-trivial and this set is not widely used. I will reconsider this if we get...
I don't think we have a convention for what to do with bridged names.