Hi.
I am using GLPK_MI to solve a MIP problem. Since the problem is NP-hard, I am happy with a suboptimal solution. I have checked GLPK_MI documentation.To stop at 10% optimality mip gap, you should pass option 'mipgap 10'; but it seems cvxpy does not pass the option to GLPK_MI.
Does anyone have a suggestion?
Here is the peace of code I call solver:
problem.solve(solver="GLPK_MI", verbose=True, options={"mipgap":"10"})
I also tired this
problem.solve(solver="GLPK_MI", verbose=True, options=["--mipgap", "10"])
Thank you,
Pooyan