You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The current MadNLP.jl dev version does solve the want_hess issue indeed, on the MWE of the last issue. It may or may not be related, but on the specific case of ModelPredictiveControl.jl (that internally uses user-registered nonlinear operator):
using ModelPredictiveControl, JuMP, MadNLP
f(x,u,_) =0.1x
h(x,_) = x
model =NonLinModel(f, h, 1, 1, 1, 1, solver=nothing)
nmpc =NonLinMPC(model, Hp=10, optim=Model(MadNLP.Optimizer))
nmpc([1])
there is no longer the want_hess error, but a new one:
good catch @franckgaga , there was an issue with the MOI wrapper in MadNLP. The problem is that eval_constraint_jacobian_product is not used inside Ipopt, so it is less thoroughly tested than the other methods.
I am opening a new issue since it may be related to another problem.
First, @frapac thanks for your work!
The current
MadNLP.jl
dev version does solve thewant_hess
issue indeed, on the MWE of the last issue. It may or may not be related, but on the specific case ofModelPredictiveControl.jl
(that internally uses user-registered nonlinear operator):there is no longer the
want_hess
error, but a new one:Originally posted by @franckgaga in #318 (comment)
The text was updated successfully, but these errors were encountered: