Setting Knitro options

I am using Knitro solver for MPEC problem. However, while solving, the solver exited before reaching the optimal decision and the information reads as EXIT: Problem appears to be unbounded. Iterate is feasible and objective magnitude > objrange. I checked the Knitro documentation how to solve objrange and they do not have example code how to fix the problem for JuMP. I appreciate any suggestion with the same experience.

This almost certainly means your problem is unbounded (the default for objrange is 1e20).

However, you can set solver-specific options like

model = Model(KNITRO.Optimizer)
set_optimizer_attribute(model, "objrange", 1e20)
1 Like