Simple disciplinary DOE example on the Sobieski SSBJ test case

from __future__ import annotations

from gemseo import configure_logger
from gemseo import create_discipline
from gemseo import create_scenario
from gemseo.problems.sobieski.core.problem import SobieskiProblem

configure_logger()
<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"])
<gemseo.algos.design_space.DesignSpace object at 0x7fea3c654430>

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"})
    INFO - 16:29:09:
    INFO - 16:29:09: *** Start DOEScenario execution ***
    INFO - 16:29:09: DOEScenario
    INFO - 16:29:09:    Disciplines: SobieskiMission
    INFO - 16:29:09:    MDO formulation: DisciplinaryOpt
    INFO - 16:29:09: Optimization problem:
    INFO - 16:29:09:    minimize -y_4(y_24, y_34)
    INFO - 16:29:09:    with respect to y_24, y_34
    INFO - 16:29:09:    over the design space:
    INFO - 16:29:09:    +------+-------------+------------+-------------+-------+
    INFO - 16:29:09:    | name | lower_bound |   value    | upper_bound | type  |
    INFO - 16:29:09:    +------+-------------+------------+-------------+-------+
    INFO - 16:29:09:    | y_24 |     0.44    | 4.15006276 |    11.13    | float |
    INFO - 16:29:09:    | y_34 |     0.44    | 1.10754577 |     1.98    | float |
    INFO - 16:29:09:    +------+-------------+------------+-------------+-------+
    INFO - 16:29:09: Solving optimization problem with algorithm lhs:
    INFO - 16:29:09: ...   0%|          | 0/30 [00:00<?, ?it]
    INFO - 16:29:09: ...   3%|▎         | 1/30 [00:00<00:00, 241.41 it/sec, obj=-1.53e+3]
    INFO - 16:29:09: ...   7%|▋         | 2/30 [00:00<00:00, 405.15 it/sec, obj=-1.66e+3]
    INFO - 16:29:09: ...  10%|█         | 3/30 [00:00<00:00, 525.84 it/sec, obj=-832]
    INFO - 16:29:09: ...  13%|█▎        | 4/30 [00:00<00:00, 620.73 it/sec, obj=-1.62e+3]
    INFO - 16:29:09: ...  17%|█▋        | 5/30 [00:00<00:00, 699.12 it/sec, obj=-994]
    INFO - 16:29:09: ...  20%|██        | 6/30 [00:00<00:00, 764.08 it/sec, obj=-601]
    INFO - 16:29:09: ...  23%|██▎       | 7/30 [00:00<00:00, 818.29 it/sec, obj=-180]
    INFO - 16:29:09: ...  27%|██▋       | 8/30 [00:00<00:00, 861.56 it/sec, obj=-755]
    INFO - 16:29:09: ...  30%|███       | 9/30 [00:00<00:00, 900.90 it/sec, obj=-691]
    INFO - 16:29:09: ...  33%|███▎      | 10/30 [00:00<00:00, 933.91 it/sec, obj=-393]
    INFO - 16:29:09: ...  37%|███▋      | 11/30 [00:00<00:00, 959.76 it/sec, obj=-362]
    INFO - 16:29:09: ...  40%|████      | 12/30 [00:00<00:00, 986.39 it/sec, obj=-748]
    INFO - 16:29:09: ...  43%|████▎     | 13/30 [00:00<00:00, 1010.04 it/sec, obj=-719]
    INFO - 16:29:09: ...  47%|████▋     | 14/30 [00:00<00:00, 1027.73 it/sec, obj=-293]
    INFO - 16:29:09: ...  50%|█████     | 15/30 [00:00<00:00, 1045.93 it/sec, obj=-931]
    INFO - 16:29:09: ...  53%|█████▎    | 16/30 [00:00<00:00, 1063.55 it/sec, obj=-264]
    INFO - 16:29:09: ...  57%|█████▋    | 17/30 [00:00<00:00, 1079.61 it/sec, obj=-1.17e+3]
    INFO - 16:29:09: ...  60%|██████    | 18/30 [00:00<00:00, 1094.34 it/sec, obj=-495]
    INFO - 16:29:09: ...  63%|██████▎   | 19/30 [00:00<00:00, 1106.00 it/sec, obj=-189]
    INFO - 16:29:09: ...  67%|██████▋   | 20/30 [00:00<00:00, 1118.26 it/sec, obj=-2.23e+3]
    INFO - 16:29:09: ...  70%|███████   | 21/30 [00:00<00:00, 1128.58 it/sec, obj=-344]
    INFO - 16:29:09: ...  73%|███████▎  | 22/30 [00:00<00:00, 1139.26 it/sec, obj=-799]
    INFO - 16:29:09: ...  77%|███████▋  | 23/30 [00:00<00:00, 1149.40 it/sec, obj=-55.9]
    INFO - 16:29:09: ...  80%|████████  | 24/30 [00:00<00:00, 1158.86 it/sec, obj=-123]
    INFO - 16:29:09: ...  83%|████████▎ | 25/30 [00:00<00:00, 1165.36 it/sec, obj=-875]
    INFO - 16:29:09: ...  87%|████████▋ | 26/30 [00:00<00:00, 1171.18 it/sec, obj=-726]
    INFO - 16:29:09: ...  90%|█████████ | 27/30 [00:00<00:00, 1176.60 it/sec, obj=-69.6]
    INFO - 16:29:09: ...  93%|█████████▎| 28/30 [00:00<00:00, 1183.58 it/sec, obj=-1.51e+3]
    INFO - 16:29:09: ...  97%|█████████▋| 29/30 [00:00<00:00, 1190.14 it/sec, obj=-1.15e+3]
    INFO - 16:29:09: ... 100%|██████████| 30/30 [00:00<00:00, 1195.33 it/sec, obj=-2.73e+3]
    INFO - 16:29:09: Optimization result:
    INFO - 16:29:09:    Optimizer info:
    INFO - 16:29:09:       Status: None
    INFO - 16:29:09:       Message: None
    INFO - 16:29:09:       Number of calls to the objective function by the optimizer: 30
    INFO - 16:29:09:    Solution:
    INFO - 16:29:09:       Objective: -2726.3660548732214
    INFO - 16:29:09:       Design space:
    INFO - 16:29:09:       +------+-------------+--------------------+-------------+-------+
    INFO - 16:29:09:       | name | lower_bound |       value        | upper_bound | type  |
    INFO - 16:29:09:       +------+-------------+--------------------+-------------+-------+
    INFO - 16:29:09:       | y_24 |     0.44    | 9.094543945649603  |    11.13    | float |
    INFO - 16:29:09:       | y_34 |     0.44    | 0.4769766573300308 |     1.98    | float |
    INFO - 16:29:09:       +------+-------------+--------------------+-------------+-------+
    INFO - 16:29:09: *** End DOEScenario execution (time: 0:00:00.037297) ***

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

Plot optimization history view

scenario.post_process("OptHistoryView", save=False, show=True)
  • Evolution of the optimization variables
  • Evolution of the objective value
  • Distance to the optimum
/home/docs/checkouts/readthedocs.org/user_builds/gemseo/envs/5.0.0/lib/python3.9/site-packages/genson/schema/strategies/base.py:42: UserWarning: Schema incompatible. Keyword 'description' has conflicting values ('The width and height of the figure in inches, e.g. ``(w, h)``.\nIf ``None``, use the :attr:`.OptPostProcessor.DEFAULT_FIG_SIZE`\nof the post-processor.' vs. 'The width and height of the figure in inches, e.g. `(w, h)`.\nIf ``None``, use the :attr:`.OptPostProcessor.DEFAULT_FIG_SIZE`\nof the post-processor.'). Using 'The width and height of the figure in inches, e.g. ``(w, h)``.\nIf ``None``, use the :attr:`.OptPostProcessor.DEFAULT_FIG_SIZE`\nof the post-processor.'
  warn(('Schema incompatible. Keyword {0!r} has conflicting '
 WARNING - 16:29:09: Failed to create Hessian approximation.
Traceback (most recent call last):
  File "/home/docs/checkouts/readthedocs.org/user_builds/gemseo/envs/5.0.0/lib/python3.9/site-packages/gemseo/post/opt_history_view.py", line 621, in _create_hessian_approx_plot
    _, diag, _, _ = SR1Approx(history).build_approximation(
  File "/home/docs/checkouts/readthedocs.org/user_builds/gemseo/envs/5.0.0/lib/python3.9/site-packages/gemseo/post/core/hessians.py", line 379, in build_approximation
    x_hist, grad_hist, _, _ = self.get_x_grad_history(
  File "/home/docs/checkouts/readthedocs.org/user_builds/gemseo/envs/5.0.0/lib/python3.9/site-packages/gemseo/post/core/hessians.py", line 170, in get_x_grad_history
    raise ValueError(
ValueError: Cannot build approximation for function: -y_4 because its gradient history is too small: 0.

<gemseo.post.opt_history_view.OptHistoryView object at 0x7fea3af1ac10>

Plot parallel coordinates

scenario.post_process(
    "ScatterPlotMatrix",
    variable_names=["y_4", "y_24", "y_34"],
    save=False,
    show=True,
)
plot sobieski doe disc example
<gemseo.post.scatter_mat.ScatterPlotMatrix object at 0x7fea46f6fc40>

Total running time of the script: ( 0 minutes 1.619 seconds)

Gallery generated by Sphinx-Gallery