Optimization History View¶

In this example, we illustrate the use of the OptHistoryView plot on the Sobieski’s SSBJ problem.

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


Import¶

The first step is to import some high-level functions and a method to get the design space.

configure_logger()

<RootLogger root (INFO)>


Description¶

The OptHistoryView post-processing creates a series of plots:

• The design variables history - This graph shows the normalized values of the design variables, the $$y$$ axis is the index of the inputs in the vector; and the $$x$$ axis represents the iterations.

• The objective function history - It shows the evolution of the objective value during the optimization.

• The distance to the best design variables - Plots the vector $$log( ||x-x^*|| )$$ in log scale.

• The history of the Hessian approximation of the objective - Plots an approximation of the second order derivatives of the objective function $$\frac{\partial^2 f(x)}{\partial x^2}$$, which is a measure of the sensitivity of the function with respect to the design variables, and of the anisotropy of the problem (differences of curvatures in the design space).

• The inequality constraint history - Portrays the evolution of the values of the constraints. The inequality constraints must be non-positive, that is why the plot must be green or white for satisfied constraints (white = active, red = violated). For an IDF formulation, an additional plot is created to track the equality constraint history.

Create disciplines¶

At this point we instantiate the disciplines of Sobieski’s SSBJ problem: Propulsion, Aerodynamics, Structure and Mission

disciplines = create_discipline(
[
"SobieskiPropulsion",
"SobieskiAerodynamics",
"SobieskiStructure",
"SobieskiMission",
]
)


Create design space¶

We also read the design space from the SobieskiProblem.

design_space = SobieskiProblem().design_space


Create and execute scenario¶

The next step is to build an MDO scenario in order to maximize the range, encoded ‘y_4’, with respect to the design parameters, while satisfying the inequality constraints ‘g_1’, ‘g_2’ and ‘g_3’. We can use the MDF formulation, the SLSQP optimization algorithm and a maximum number of iterations equal to 100.

scenario = create_scenario(
disciplines,
formulation="MDF",
objective_name="y_4",
maximize_objective=True,
design_space=design_space,
)
scenario.set_differentiation_method()
for constraint in ["g_1", "g_2", "g_3"]:
scenario.execute({"algo": "SLSQP", "max_iter": 100})

    INFO - 08:27:44:
INFO - 08:27:44: *** Start MDOScenario execution ***
INFO - 08:27:44: MDOScenario
INFO - 08:27:44:    Disciplines: SobieskiAerodynamics SobieskiMission SobieskiPropulsion SobieskiStructure
INFO - 08:27:44:    MDO formulation: MDF
INFO - 08:27:44: Optimization problem:
INFO - 08:27:44:    minimize -y_4(x_shared, x_1, x_2, x_3)
INFO - 08:27:44:    with respect to x_1, x_2, x_3, x_shared
INFO - 08:27:44:    subject to constraints:
INFO - 08:27:44:       g_1(x_shared, x_1, x_2, x_3) <= 0.0
INFO - 08:27:44:       g_2(x_shared, x_1, x_2, x_3) <= 0.0
INFO - 08:27:44:       g_3(x_shared, x_1, x_2, x_3) <= 0.0
INFO - 08:27:44:    over the design space:
INFO - 08:27:44:    +-------------+-------------+-------+-------------+-------+
INFO - 08:27:44:    | name        | lower_bound | value | upper_bound | type  |
INFO - 08:27:44:    +-------------+-------------+-------+-------------+-------+
INFO - 08:27:44:    | x_shared[0] |     0.01    |  0.05 |     0.09    | float |
INFO - 08:27:44:    | x_shared[1] |    30000    | 45000 |    60000    | float |
INFO - 08:27:44:    | x_shared[2] |     1.4     |  1.6  |     1.8     | float |
INFO - 08:27:44:    | x_shared[3] |     2.5     |  5.5  |     8.5     | float |
INFO - 08:27:44:    | x_shared[4] |      40     |   55  |      70     | float |
INFO - 08:27:44:    | x_shared[5] |     500     |  1000 |     1500    | float |
INFO - 08:27:44:    | x_1[0]      |     0.1     |  0.25 |     0.4     | float |
INFO - 08:27:44:    | x_1[1]      |     0.75    |   1   |     1.25    | float |
INFO - 08:27:44:    | x_2         |     0.75    |   1   |     1.25    | float |
INFO - 08:27:44:    | x_3         |     0.1     |  0.5  |      1      | float |
INFO - 08:27:44:    +-------------+-------------+-------+-------------+-------+
INFO - 08:27:44: Solving optimization problem with algorithm SLSQP:
INFO - 08:27:44: ...   0%|          | 0/100 [00:00<?, ?it]
INFO - 08:27:44: ...   1%|          | 1/100 [00:00<00:09, 10.16 it/sec, obj=-536]
INFO - 08:27:44: ...   2%|▏         | 2/100 [00:00<00:13,  7.49 it/sec, obj=-2.12e+3]
WARNING - 08:27:45: MDAJacobi has reached its maximum number of iterations but the normed residual 1.7130677857005655e-05 is still above the tolerance 1e-06.
INFO - 08:27:45: ...   3%|▎         | 3/100 [00:00<00:15,  6.28 it/sec, obj=-3.75e+3]
INFO - 08:27:45: ...   4%|▍         | 4/100 [00:00<00:16,  6.00 it/sec, obj=-3.96e+3]
INFO - 08:27:45: ...   5%|▌         | 5/100 [00:00<00:16,  5.84 it/sec, obj=-3.96e+3]
INFO - 08:27:45: Optimization result:
INFO - 08:27:45:    Optimizer info:
INFO - 08:27:45:       Status: 8
INFO - 08:27:45:       Message: Positive directional derivative for linesearch
INFO - 08:27:45:       Number of calls to the objective function by the optimizer: 6
INFO - 08:27:45:    Solution:
INFO - 08:27:45:       The solution is feasible.
INFO - 08:27:45:       Objective: -3963.408265187933
INFO - 08:27:45:       Standardized constraints:
INFO - 08:27:45:          g_1 = [-0.01806104 -0.03334642 -0.04424946 -0.0518346  -0.05732607 -0.13720865
INFO - 08:27:45:  -0.10279135]
INFO - 08:27:45:          g_2 = 3.333278582928756e-06
INFO - 08:27:45:          g_3 = [-7.67181773e-01 -2.32818227e-01  8.30379541e-07 -1.83255000e-01]
INFO - 08:27:45:       Design space:
INFO - 08:27:45:       +-------------+-------------+---------------------+-------------+-------+
INFO - 08:27:45:       | name        | lower_bound |        value        | upper_bound | type  |
INFO - 08:27:45:       +-------------+-------------+---------------------+-------------+-------+
INFO - 08:27:45:       | x_shared[0] |     0.01    | 0.06000083331964572 |     0.09    | float |
INFO - 08:27:45:       | x_shared[1] |    30000    |        60000        |    60000    | float |
INFO - 08:27:45:       | x_shared[2] |     1.4     |         1.4         |     1.8     | float |
INFO - 08:27:45:       | x_shared[3] |     2.5     |         2.5         |     8.5     | float |
INFO - 08:27:45:       | x_shared[4] |      40     |          70         |      70     | float |
INFO - 08:27:45:       | x_shared[5] |     500     |         1500        |     1500    | float |
INFO - 08:27:45:       | x_1[0]      |     0.1     |         0.4         |     0.4     | float |
INFO - 08:27:45:       | x_1[1]      |     0.75    |         0.75        |     1.25    | float |
INFO - 08:27:45:       | x_2         |     0.75    |         0.75        |     1.25    | float |
INFO - 08:27:45:       | x_3         |     0.1     |  0.1562448753887276 |      1      | float |
INFO - 08:27:45:       +-------------+-------------+---------------------+-------------+-------+
INFO - 08:27:45: *** End MDOScenario execution (time: 0:00:00.972598) ***

{'max_iter': 100, 'algo': 'SLSQP'}


Post-process scenario¶

Lastly, we post-process the scenario by means of the OptHistoryView plot which plots the history of optimization for both objective function, constraints, design parameters and distance to the optimum.

Tip

Each post-processing method requires different inputs and offers a variety of customization options. Use the high-level function get_post_processing_options_schema() to print a table with the options for any post-processing algorithm. Or refer to our dedicated page: Post-processing algorithms.

scenario.post_process(
"OptHistoryView", save=False, show=True, variable_names=["x_2", "x_1"]
)

<gemseo.post.opt_history_view.OptHistoryView object at 0x7f0cb47c2be0>


Total running time of the script: (0 minutes 2.040 seconds)

Gallery generated by Sphinx-Gallery