Hypatia.jl
Hypatia.jl copied to clipboard
interior point solver for general convex conic optimization problems
solves auxiliary conic problems if necessary, with some zeroed-out b, c, or h. see https://epubs.siam.org/doi/pdf/10.1137/15M1049415
`MOI.ReductionCertificate`, `MOI.NearlyReductionCertificate` see this [reference](https://epubs.siam.org/doi/pdf/10.1137/15M1049415?casa_token=oyqqEGqc7LkAAAAA:R6F5BD6-9EluZp7TCIDKGMrXit_ifnrJ7UOprNbwOhPX0dY_7g1vbgC9uCYOuzu0scfVTQJkHJo) I think there should be (optional) MOI.Test tests for facial reduction certificates. @blegat what do you think?
Can also probably do it without needing values for both primal and dual variables Particular use case is when just changing the b-vector in a mixed integer convex OA algorithm....
With this PR, when the user uses the sampling basis, e.g. with ```julia @constraint(model, p in SOSCone(), zero_basis = BoxSampling(-ones(nvariables(p)), ones(nvariables(p)))) ``` the constraint will be transformed to `Hypatia.WSOSInterpNonnegativeCone` and...
...as the parameter was removed in #825