It would be nice for NLPModels to define a standard API for the data we'd need for sensitivity analysis. In particular:
- Partial second derivatives of the lagrangian w.r.t. primal solution and parameters (e.g.
NLPModels.hess_param(nlp, x, y))
- Partial derivatives of the constraints w.r.t. parameters (e.g.
NLPModels.jac_param(nlp, x)?)
It would be nice for NLPModels to define a standard API for the data we'd need for sensitivity analysis. In particular:
NLPModels.hess_param(nlp, x, y))NLPModels.jac_param(nlp, x)?)