Simple disciplinary DOE example on the Sobieski SSBJ test case

from gemseo.api import configure_logger
from gemseo.api import create_discipline
from gemseo.api import create_scenario
from gemseo.problems.sobieski.core.problem import SobieskiProblem
from matplotlib import pyplot as plt

configure_logger()

Out:

<RootLogger root (INFO)>

Instantiate the discipline

discipline = create_discipline("SobieskiMission")

Create the design space

design_space = SobieskiProblem().design_space
design_space.filter(["y_24", "y_34"])

Out:

<gemseo.algos.design_space.DesignSpace object at 0x7fdbe0d70e50>

Create the scenario

Build scenario which links the disciplines with the formulation and The DOE algorithm.

scenario = create_scenario(
    [discipline],
    formulation="DisciplinaryOpt",
    objective_name="y_4",
    design_space=design_space,
    maximize_objective=True,
    scenario_type="DOE",
)

Execute the scenario

Here we use a latin hypercube sampling algorithm with 30 samples.

scenario.execute({"n_samples": 30, "algo": "lhs"})

Out:

    INFO - 10:06:56:
    INFO - 10:06:56: *** Start DOEScenario execution ***
    INFO - 10:06:56: DOEScenario
    INFO - 10:06:56:    Disciplines: SobieskiMission
    INFO - 10:06:56:    MDO formulation: DisciplinaryOpt
    INFO - 10:06:56: Optimization problem:
    INFO - 10:06:56:    minimize -y_4(y_24, y_34)
    INFO - 10:06:56:    with respect to y_24, y_34
    INFO - 10:06:56:    over the design space:
    INFO - 10:06:56:    +------+-------------+------------+-------------+-------+
    INFO - 10:06:56:    | name | lower_bound |   value    | upper_bound | type  |
    INFO - 10:06:56:    +------+-------------+------------+-------------+-------+
    INFO - 10:06:56:    | y_24 |     0.44    | 4.15006276 |    11.13    | float |
    INFO - 10:06:56:    | y_34 |     0.44    | 1.10754577 |     1.98    | float |
    INFO - 10:06:56:    +------+-------------+------------+-------------+-------+
    INFO - 10:06:56: Solving optimization problem with algorithm lhs:
    INFO - 10:06:56: ...   0%|          | 0/30 [00:00<?, ?it]
    INFO - 10:06:56: ... 100%|██████████| 30/30 [00:00<00:00, 1761.01 it/sec, obj=-2.73e+3]
    INFO - 10:06:56: Optimization result:
    INFO - 10:06:56:    Optimizer info:
    INFO - 10:06:56:       Status: None
    INFO - 10:06:56:       Message: None
    INFO - 10:06:56:       Number of calls to the objective function by the optimizer: 30
    INFO - 10:06:56:    Solution:
    INFO - 10:06:56:       Objective: -2726.3660548732214
    INFO - 10:06:56:       Design space:
    INFO - 10:06:56:       +------+-------------+--------------------+-------------+-------+
    INFO - 10:06:56:       | name | lower_bound |       value        | upper_bound | type  |
    INFO - 10:06:56:       +------+-------------+--------------------+-------------+-------+
    INFO - 10:06:56:       | y_24 |     0.44    | 9.094543945649603  |    11.13    | float |
    INFO - 10:06:56:       | y_34 |     0.44    | 0.4769766573300308 |     1.98    | float |
    INFO - 10:06:56:       +------+-------------+--------------------+-------------+-------+
    INFO - 10:06:56: *** End DOEScenario execution (time: 0:00:00.025205) ***

{'eval_jac': False, 'algo': 'lhs', 'n_samples': 30}

Plot optimization history view

scenario.post_process("OptHistoryView", save=False, show=False)

• 
• 
• 

Out:

<gemseo.post.opt_history_view.OptHistoryView object at 0x7fdbe0b37940>

Plot parallel coordinates

scenario.post_process(
    "ScatterPlotMatrix", save=False, show=False, variable_names=["y_4", "y_24", "y_34"]
)
# Workaround for HTML rendering, instead of ``show=True``
plt.show()