Note
Go to the end to download the full example code.
Execute a scenario using a DOE#
from __future__ import annotations
from gemseo import create_discipline
from gemseo import create_scenario
from gemseo.problems.mdo.sobieski.core.design_space import SobieskiDesignSpace
Instantiate the discipline#
discipline = create_discipline("SobieskiMission")
Create the design space#
design_space = SobieskiDesignSpace()
design_space.filter(["y_24", "y_34"])
Create the scenario#
Build scenario which links the disciplines with the formulation and The DOE algorithm.
scenario = create_scenario(
[discipline],
"y_4",
design_space,
maximize_objective=True,
scenario_type="DOE",
formulation_name="DisciplinaryOpt",
)
Execute the scenario#
Here we use a latin hypercube sampling algorithm with 30 samples.
scenario.execute(algo_name="PYDOE_LHS", n_samples=30)
INFO - 16:17:37: *** Start DOEScenario execution ***
INFO - 16:17:37: DOEScenario
INFO - 16:17:37: Disciplines: SobieskiMission
INFO - 16:17:37: MDO formulation: DisciplinaryOpt
INFO - 16:17:37: Optimization problem:
INFO - 16:17:37: minimize -y_4(y_24, y_34)
INFO - 16:17:37: with respect to y_24, y_34
INFO - 16:17:37: over the design space:
INFO - 16:17:37: +------+-------------+------------+-------------+-------+
INFO - 16:17:37: | Name | Lower bound | Value | Upper bound | Type |
INFO - 16:17:37: +------+-------------+------------+-------------+-------+
INFO - 16:17:37: | y_24 | 0.44 | 4.15006276 | 11.13 | float |
INFO - 16:17:37: | y_34 | 0.44 | 1.10754577 | 1.98 | float |
INFO - 16:17:37: +------+-------------+------------+-------------+-------+
INFO - 16:17:37: Solving optimization problem with algorithm PYDOE_LHS:
INFO - 16:17:37: 3%|▎ | 1/30 [00:00<00:00, 340.36 it/sec, feas=True, obj=-1.53e+3]
INFO - 16:17:37: 7%|▋ | 2/30 [00:00<00:00, 579.28 it/sec, feas=True, obj=-1.66e+3]
INFO - 16:17:37: 10%|█ | 3/30 [00:00<00:00, 761.95 it/sec, feas=True, obj=-832]
INFO - 16:17:37: 13%|█▎ | 4/30 [00:00<00:00, 866.59 it/sec, feas=True, obj=-1.62e+3]
INFO - 16:17:37: 17%|█▋ | 5/30 [00:00<00:00, 962.61 it/sec, feas=True, obj=-994]
INFO - 16:17:37: 20%|██ | 6/30 [00:00<00:00, 1058.63 it/sec, feas=True, obj=-601]
INFO - 16:17:37: 23%|██▎ | 7/30 [00:00<00:00, 1140.02 it/sec, feas=True, obj=-180]
INFO - 16:17:37: 27%|██▋ | 8/30 [00:00<00:00, 1211.53 it/sec, feas=True, obj=-755]
INFO - 16:17:37: 30%|███ | 9/30 [00:00<00:00, 1274.18 it/sec, feas=True, obj=-691]
INFO - 16:17:37: 33%|███▎ | 10/30 [00:00<00:00, 1327.10 it/sec, feas=True, obj=-393]
INFO - 16:17:37: 37%|███▋ | 11/30 [00:00<00:00, 1374.57 it/sec, feas=True, obj=-362]
INFO - 16:17:37: 40%|████ | 12/30 [00:00<00:00, 1416.76 it/sec, feas=True, obj=-748]
INFO - 16:17:37: 43%|████▎ | 13/30 [00:00<00:00, 1448.00 it/sec, feas=True, obj=-719]
INFO - 16:17:37: 47%|████▋ | 14/30 [00:00<00:00, 1482.87 it/sec, feas=True, obj=-293]
INFO - 16:17:37: 50%|█████ | 15/30 [00:00<00:00, 1516.63 it/sec, feas=True, obj=-931]
INFO - 16:17:37: 53%|█████▎ | 16/30 [00:00<00:00, 1540.50 it/sec, feas=True, obj=-264]
INFO - 16:17:37: 57%|█████▋ | 17/30 [00:00<00:00, 1567.58 it/sec, feas=True, obj=-1.17e+3]
INFO - 16:17:37: 60%|██████ | 18/30 [00:00<00:00, 1590.29 it/sec, feas=True, obj=-495]
INFO - 16:17:37: 63%|██████▎ | 19/30 [00:00<00:00, 1613.81 it/sec, feas=True, obj=-189]
INFO - 16:17:37: 67%|██████▋ | 20/30 [00:00<00:00, 1634.89 it/sec, feas=True, obj=-2.23e+3]
INFO - 16:17:37: 70%|███████ | 21/30 [00:00<00:00, 1652.63 it/sec, feas=True, obj=-344]
INFO - 16:17:37: 73%|███████▎ | 22/30 [00:00<00:00, 1666.51 it/sec, feas=True, obj=-799]
INFO - 16:17:37: 77%|███████▋ | 23/30 [00:00<00:00, 1683.37 it/sec, feas=True, obj=-55.9]
INFO - 16:17:37: 80%|████████ | 24/30 [00:00<00:00, 1700.97 it/sec, feas=True, obj=-123]
INFO - 16:17:37: 83%|████████▎ | 25/30 [00:00<00:00, 1716.81 it/sec, feas=True, obj=-875]
INFO - 16:17:37: 87%|████████▋ | 26/30 [00:00<00:00, 1730.16 it/sec, feas=True, obj=-726]
INFO - 16:17:37: 90%|█████████ | 27/30 [00:00<00:00, 1742.33 it/sec, feas=True, obj=-69.6]
INFO - 16:17:37: 93%|█████████▎| 28/30 [00:00<00:00, 1750.83 it/sec, feas=True, obj=-1.51e+3]
INFO - 16:17:37: 97%|█████████▋| 29/30 [00:00<00:00, 1761.65 it/sec, feas=True, obj=-1.15e+3]
INFO - 16:17:37: 100%|██████████| 30/30 [00:00<00:00, 1753.03 it/sec, feas=True, obj=-2.73e+3]
INFO - 16:17:37: Optimization result:
INFO - 16:17:37: Optimizer info:
INFO - 16:17:37: Status: None
INFO - 16:17:37: Message: None
INFO - 16:17:37: Solution:
INFO - 16:17:37: Objective: -2726.3660548732214
INFO - 16:17:37: Design space:
INFO - 16:17:37: +------+-------------+--------------------+-------------+-------+
INFO - 16:17:37: | Name | Lower bound | Value | Upper bound | Type |
INFO - 16:17:37: +------+-------------+--------------------+-------------+-------+
INFO - 16:17:37: | y_24 | 0.44 | 9.094543945649603 | 11.13 | float |
INFO - 16:17:37: | y_34 | 0.44 | 0.4769766573300308 | 1.98 | float |
INFO - 16:17:37: +------+-------------+--------------------+-------------+-------+
INFO - 16:17:37: *** End DOEScenario execution ***
Note that both the formulation settings passed to create_scenario() and the
algorithm settings passed to execute()
can be provided via a Pydantic model.
For more information, see Formulation Settings and Algorithm Settings.
Plot optimization history view#
scenario.post_process(post_name="OptHistoryView", save=False, show=True)
<gemseo.post.opt_history_view.OptHistoryView object at 0x7b1d50f84530>
Note that post-processor settings passed to post_process() can be
provided via a Pydantic model (see the example below). For more information,
see Post-processor Settings.
Plot scatter plot matrix#
from gemseo.settings.post import ScatterPlotMatrix_Settings # noqa: E402
settings_model = ScatterPlotMatrix_Settings(
variable_names=["y_4", "y_24", "y_34"],
save=False,
show=True,
)
scenario.post_process(settings_model)
<gemseo.post.scatter_plot_matrix.ScatterPlotMatrix object at 0x7b1d3196d8e0>
Total running time of the script: (0 minutes 0.787 seconds)