Scenario

from gemseo import create_discipline
from gemseo import create_scenario
from gemseo import read_design_space

Instantiate an MDO or DOE scenario:

discipline_names = ["disc1", "disc2", "disc3"]
disciplines = create_discipline(discipline_names, "ext_path")
design_space = read_design_space("file.csv")
scenario_type = "MDO"  # or 'DOE'
scenario = create_scenario(
    disciplines,
    formulation_name="MDF",
    objective_name="obj",
    design_space=design_space,
    name="my_scenario",
    scenario_type=scenario_type,
    **formulation_options,
)
scenario.add_constraint("cstr1")  # =0
scenario.add_constraint("cstr2", value=1.0)  # =1
scenario.add_constraint("cstr3", constraint_type="ineq")  # <=0
scenario.add_constraint("cstr4", constraint_type="ineq", value=1.0)  # <=1
scenario.add_constraint("cstr5", constraint_type="ineq", positive=True)  # >=0
scenario.add_constraint("cstr6", constraint_type="ineq", positive=True, value=1.0)  # >=1
scenario.xdsmize()  # Build the XDSM graph to check it.

Execute the scenario:

scenario.execute(algo_name="SLSQP", max_iter=50, xtol_rel=1e-3)  # if MDO
scenario.execute(algo_name="LHS", n_samples=30)  # if DOE
optimum = scenario.get_optimum()

Save the optimisation history:

optimum_result = scenario.save_optimization_history("file.h5")