Note
Click here to download the full example code
Create a DOE Scenario¶
from __future__ import annotations
from gemseo.api import configure_logger
from gemseo.api import create_design_space
from gemseo.api import create_discipline
from gemseo.api import create_scenario
from gemseo.api import get_available_doe_algorithms
from gemseo.api import get_available_post_processings
configure_logger()
<RootLogger root (INFO)>
Let \((P)\) be a simple optimization problem:
In this example, we will see how to use GEMSEO to solve this problem \((P)\) by means of a Design Of Experiments (DOE)
Define the discipline¶
Firstly, by means of the create_discipline()
API function,
we create an MDODiscipline
of AnalyticDiscipline
type
from a Python function:
expressions = {"y": "x1+x2"}
discipline = create_discipline("AnalyticDiscipline", expressions=expressions)
Now, we want to minimize this MDODiscipline
over a design of experiments (DOE).
Define the design space¶
For that, by means of the create_design_space()
API function,
we define the DesignSpace
\([-5, 5]\times[-5, 5]\)
by using its DesignSpace.add_variable()
method.
design_space = create_design_space()
design_space.add_variable("x1", l_b=-5, u_b=5, var_type="integer")
design_space.add_variable("x2", l_b=-5, u_b=5, var_type="integer")
Define the DOE scenario¶
Then, by means of the create_scenario()
API function,
we define a DOEScenario
from the MDODiscipline
and the DesignSpace
defined above:
scenario = create_scenario(
discipline, "DisciplinaryOpt", "y", design_space, scenario_type="DOE"
)
Execute the DOE scenario¶
Lastly, we solve the OptimizationProblem
included in the
DOEScenario
defined above by minimizing the objective function
over a design of experiments included in the DesignSpace
.
Precisely, we choose a full factorial design of size \(11^2\):
scenario.execute({"algo": "fullfact", "n_samples": 11**2})
INFO - 13:28:00:
INFO - 13:28:00: *** Start DOEScenario execution ***
INFO - 13:28:00: DOEScenario
INFO - 13:28:00: Disciplines: AnalyticDiscipline
INFO - 13:28:00: MDO formulation: DisciplinaryOpt
INFO - 13:28:00: Optimization problem:
INFO - 13:28:00: minimize y(x1, x2)
INFO - 13:28:00: with respect to x1, x2
INFO - 13:28:00: over the design space:
INFO - 13:28:00: +------+-------------+-------+-------------+---------+
INFO - 13:28:00: | name | lower_bound | value | upper_bound | type |
INFO - 13:28:00: +------+-------------+-------+-------------+---------+
INFO - 13:28:00: | x1 | -5 | None | 5 | integer |
INFO - 13:28:00: | x2 | -5 | None | 5 | integer |
INFO - 13:28:00: +------+-------------+-------+-------------+---------+
INFO - 13:28:00: Solving optimization problem with algorithm fullfact:
INFO - 13:28:00: ... 0%| | 0/121 [00:00<?, ?it]
INFO - 13:28:00: ... 1%| | 1/121 [00:00<00:00, 388.18 it/sec, obj=-10]
INFO - 13:28:00: ... 2%|▏ | 2/121 [00:00<00:00, 621.98 it/sec, obj=-9]
INFO - 13:28:00: ... 2%|▏ | 3/121 [00:00<00:00, 777.44 it/sec, obj=-8]
INFO - 13:28:00: ... 3%|▎ | 4/121 [00:00<00:00, 892.74 it/sec, obj=-7]
INFO - 13:28:00: ... 4%|▍ | 5/121 [00:00<00:00, 982.00 it/sec, obj=-6]
INFO - 13:28:00: ... 5%|▍ | 6/121 [00:00<00:00, 1052.30 it/sec, obj=-5]
INFO - 13:28:00: ... 6%|▌ | 7/121 [00:00<00:00, 1100.08 it/sec, obj=-4]
INFO - 13:28:00: ... 7%|▋ | 8/121 [00:00<00:00, 1146.81 it/sec, obj=-3]
INFO - 13:28:00: ... 7%|▋ | 9/121 [00:00<00:00, 1183.09 it/sec, obj=-2]
INFO - 13:28:00: ... 8%|▊ | 10/121 [00:00<00:00, 1216.55 it/sec, obj=-1]
INFO - 13:28:00: ... 9%|▉ | 11/121 [00:00<00:00, 1246.28 it/sec, obj=0]
INFO - 13:28:00: ... 10%|▉ | 12/121 [00:00<00:00, 1272.03 it/sec, obj=-9]
INFO - 13:28:00: ... 11%|█ | 13/121 [00:00<00:00, 1290.77 it/sec, obj=-8]
INFO - 13:28:00: ... 12%|█▏ | 14/121 [00:00<00:00, 1310.78 it/sec, obj=-7]
INFO - 13:28:00: ... 12%|█▏ | 15/121 [00:00<00:00, 1328.80 it/sec, obj=-6]
INFO - 13:28:00: ... 13%|█▎ | 16/121 [00:00<00:00, 1343.52 it/sec, obj=-5]
INFO - 13:28:00: ... 14%|█▍ | 17/121 [00:00<00:00, 1358.28 it/sec, obj=-4]
INFO - 13:28:00: ... 15%|█▍ | 18/121 [00:00<00:00, 1371.06 it/sec, obj=-3]
INFO - 13:28:00: ... 16%|█▌ | 19/121 [00:00<00:00, 1383.49 it/sec, obj=-2]
INFO - 13:28:00: ... 17%|█▋ | 20/121 [00:00<00:00, 1391.31 it/sec, obj=-1]
INFO - 13:28:00: ... 17%|█▋ | 21/121 [00:00<00:00, 1402.02 it/sec, obj=0]
INFO - 13:28:00: ... 18%|█▊ | 22/121 [00:00<00:00, 1410.58 it/sec, obj=1]
INFO - 13:28:00: ... 19%|█▉ | 23/121 [00:00<00:00, 1419.00 it/sec, obj=-8]
INFO - 13:28:00: ... 20%|█▉ | 24/121 [00:00<00:00, 1427.40 it/sec, obj=-7]
INFO - 13:28:00: ... 21%|██ | 25/121 [00:00<00:00, 1435.23 it/sec, obj=-6]
INFO - 13:28:00: ... 21%|██▏ | 26/121 [00:00<00:00, 1440.47 it/sec, obj=-5]
INFO - 13:28:00: ... 22%|██▏ | 27/121 [00:00<00:00, 1447.03 it/sec, obj=-4]
INFO - 13:28:00: ... 23%|██▎ | 28/121 [00:00<00:00, 1453.49 it/sec, obj=-3]
INFO - 13:28:00: ... 24%|██▍ | 29/121 [00:00<00:00, 1458.17 it/sec, obj=-2]
INFO - 13:28:00: ... 25%|██▍ | 30/121 [00:00<00:00, 1463.79 it/sec, obj=-1]
INFO - 13:28:00: ... 26%|██▌ | 31/121 [00:00<00:00, 1469.21 it/sec, obj=0]
INFO - 13:28:00: ... 26%|██▋ | 32/121 [00:00<00:00, 1474.21 it/sec, obj=1]
INFO - 13:28:00: ... 27%|██▋ | 33/121 [00:00<00:00, 1477.01 it/sec, obj=2]
INFO - 13:28:00: ... 28%|██▊ | 34/121 [00:00<00:00, 1481.50 it/sec, obj=-7]
INFO - 13:28:00: ... 29%|██▉ | 35/121 [00:00<00:00, 1484.72 it/sec, obj=-6]
INFO - 13:28:00: ... 30%|██▉ | 36/121 [00:00<00:00, 1488.46 it/sec, obj=-5]
INFO - 13:28:00: ... 31%|███ | 37/121 [00:00<00:00, 1492.12 it/sec, obj=-4]
INFO - 13:28:00: ... 31%|███▏ | 38/121 [00:00<00:00, 1495.69 it/sec, obj=-3]
INFO - 13:28:00: ... 32%|███▏ | 39/121 [00:00<00:00, 1497.39 it/sec, obj=-2]
INFO - 13:28:00: ... 33%|███▎ | 40/121 [00:00<00:00, 1500.48 it/sec, obj=-1]
INFO - 13:28:00: ... 34%|███▍ | 41/121 [00:00<00:00, 1503.94 it/sec, obj=0]
INFO - 13:28:00: ... 35%|███▍ | 42/121 [00:00<00:00, 1506.06 it/sec, obj=1]
INFO - 13:28:00: ... 36%|███▌ | 43/121 [00:00<00:00, 1508.96 it/sec, obj=2]
INFO - 13:28:00: ... 36%|███▋ | 44/121 [00:00<00:00, 1512.07 it/sec, obj=3]
INFO - 13:28:00: ... 37%|███▋ | 45/121 [00:00<00:00, 1514.97 it/sec, obj=-6]
INFO - 13:28:00: ... 38%|███▊ | 46/121 [00:00<00:00, 1516.15 it/sec, obj=-5]
INFO - 13:28:00: ... 39%|███▉ | 47/121 [00:00<00:00, 1516.23 it/sec, obj=-4]
INFO - 13:28:00: ... 40%|███▉ | 48/121 [00:00<00:00, 1514.79 it/sec, obj=-3]
INFO - 13:28:00: ... 40%|████ | 49/121 [00:00<00:00, 1516.75 it/sec, obj=-2]
INFO - 13:28:00: ... 41%|████▏ | 50/121 [00:00<00:00, 1518.92 it/sec, obj=-1]
INFO - 13:28:00: ... 42%|████▏ | 51/121 [00:00<00:00, 1521.13 it/sec, obj=0]
INFO - 13:28:00: ... 43%|████▎ | 52/121 [00:00<00:00, 1521.88 it/sec, obj=1]
INFO - 13:28:00: ... 44%|████▍ | 53/121 [00:00<00:00, 1523.86 it/sec, obj=2]
INFO - 13:28:00: ... 45%|████▍ | 54/121 [00:00<00:00, 1523.53 it/sec, obj=3]
INFO - 13:28:00: ... 45%|████▌ | 55/121 [00:00<00:00, 1524.99 it/sec, obj=4]
INFO - 13:28:00: ... 46%|████▋ | 56/121 [00:00<00:00, 1526.76 it/sec, obj=-5]
INFO - 13:28:00: ... 47%|████▋ | 57/121 [00:00<00:00, 1526.76 it/sec, obj=-4]
INFO - 13:28:00: ... 48%|████▊ | 58/121 [00:00<00:00, 1525.97 it/sec, obj=-3]
INFO - 13:28:00: ... 49%|████▉ | 59/121 [00:00<00:00, 1527.28 it/sec, obj=-2]
INFO - 13:28:00: ... 50%|████▉ | 60/121 [00:00<00:00, 1528.91 it/sec, obj=-1]
INFO - 13:28:00: ... 50%|█████ | 61/121 [00:00<00:00, 1529.80 it/sec, obj=0]
INFO - 13:28:00: ... 51%|█████ | 62/121 [00:00<00:00, 1531.42 it/sec, obj=1]
INFO - 13:28:00: ... 52%|█████▏ | 63/121 [00:00<00:00, 1533.09 it/sec, obj=2]
INFO - 13:28:00: ... 53%|█████▎ | 64/121 [00:00<00:00, 1529.42 it/sec, obj=3]
INFO - 13:28:00: ... 54%|█████▎ | 65/121 [00:00<00:00, 1530.89 it/sec, obj=4]
INFO - 13:28:00: ... 55%|█████▍ | 66/121 [00:00<00:00, 1532.29 it/sec, obj=5]
INFO - 13:28:00: ... 55%|█████▌ | 67/121 [00:00<00:00, 1533.24 it/sec, obj=-4]
INFO - 13:28:00: ... 56%|█████▌ | 68/121 [00:00<00:00, 1534.66 it/sec, obj=-3]
INFO - 13:28:00: ... 57%|█████▋ | 69/121 [00:00<00:00, 1536.15 it/sec, obj=-2]
INFO - 13:28:00: ... 58%|█████▊ | 70/121 [00:00<00:00, 1537.64 it/sec, obj=-1]
INFO - 13:28:00: ... 59%|█████▊ | 71/121 [00:00<00:00, 1538.10 it/sec, obj=0]
INFO - 13:28:00: ... 60%|█████▉ | 72/121 [00:00<00:00, 1539.34 it/sec, obj=1]
INFO - 13:28:00: ... 60%|██████ | 73/121 [00:00<00:00, 1539.96 it/sec, obj=2]
INFO - 13:28:00: ... 61%|██████ | 74/121 [00:00<00:00, 1540.51 it/sec, obj=3]
INFO - 13:28:00: ... 62%|██████▏ | 75/121 [00:00<00:00, 1541.77 it/sec, obj=4]
INFO - 13:28:00: ... 63%|██████▎ | 76/121 [00:00<00:00, 1542.93 it/sec, obj=5]
INFO - 13:28:00: ... 64%|██████▎ | 77/121 [00:00<00:00, 1543.29 it/sec, obj=6]
INFO - 13:28:00: ... 64%|██████▍ | 78/121 [00:00<00:00, 1544.32 it/sec, obj=-3]
INFO - 13:28:00: ... 65%|██████▌ | 79/121 [00:00<00:00, 1545.63 it/sec, obj=-2]
INFO - 13:28:00: ... 66%|██████▌ | 80/121 [00:00<00:00, 1546.34 it/sec, obj=-1]
INFO - 13:28:00: ... 67%|██████▋ | 81/121 [00:00<00:00, 1547.50 it/sec, obj=0]
INFO - 13:28:00: ... 68%|██████▊ | 82/121 [00:00<00:00, 1548.74 it/sec, obj=1]
INFO - 13:28:00: ... 69%|██████▊ | 83/121 [00:00<00:00, 1549.98 it/sec, obj=2]
INFO - 13:28:00: ... 69%|██████▉ | 84/121 [00:00<00:00, 1550.37 it/sec, obj=3]
INFO - 13:28:00: ... 70%|███████ | 85/121 [00:00<00:00, 1551.45 it/sec, obj=4]
INFO - 13:28:00: ... 71%|███████ | 86/121 [00:00<00:00, 1552.61 it/sec, obj=5]
INFO - 13:28:00: ... 72%|███████▏ | 87/121 [00:00<00:00, 1553.06 it/sec, obj=6]
INFO - 13:28:00: ... 73%|███████▎ | 88/121 [00:00<00:00, 1554.13 it/sec, obj=7]
INFO - 13:28:00: ... 74%|███████▎ | 89/121 [00:00<00:00, 1555.20 it/sec, obj=-2]
INFO - 13:28:00: ... 74%|███████▍ | 90/121 [00:00<00:00, 1555.55 it/sec, obj=-1]
INFO - 13:28:00: ... 75%|███████▌ | 91/121 [00:00<00:00, 1556.37 it/sec, obj=0]
INFO - 13:28:00: ... 76%|███████▌ | 92/121 [00:00<00:00, 1557.38 it/sec, obj=1]
INFO - 13:28:00: ... 77%|███████▋ | 93/121 [00:00<00:00, 1557.89 it/sec, obj=2]
INFO - 13:28:00: ... 78%|███████▊ | 94/121 [00:00<00:00, 1558.82 it/sec, obj=3]
INFO - 13:28:00: ... 79%|███████▊ | 95/121 [00:00<00:00, 1559.81 it/sec, obj=4]
INFO - 13:28:00: ... 79%|███████▉ | 96/121 [00:00<00:00, 1560.67 it/sec, obj=5]
INFO - 13:28:00: ... 80%|████████ | 97/121 [00:00<00:00, 1559.56 it/sec, obj=6]
INFO - 13:28:00: ... 81%|████████ | 98/121 [00:00<00:00, 1559.57 it/sec, obj=7]
INFO - 13:28:00: ... 82%|████████▏ | 99/121 [00:00<00:00, 1560.02 it/sec, obj=8]
INFO - 13:28:00: ... 83%|████████▎ | 100/121 [00:00<00:00, 1560.64 it/sec, obj=-1]
INFO - 13:28:00: ... 83%|████████▎ | 101/121 [00:00<00:00, 1561.43 it/sec, obj=0]
INFO - 13:28:00: ... 84%|████████▍ | 102/121 [00:00<00:00, 1562.20 it/sec, obj=1]
INFO - 13:28:00: ... 85%|████████▌ | 103/121 [00:00<00:00, 1562.25 it/sec, obj=2]
INFO - 13:28:00: ... 86%|████████▌ | 104/121 [00:00<00:00, 1562.86 it/sec, obj=3]
INFO - 13:28:00: ... 87%|████████▋ | 105/121 [00:00<00:00, 1563.59 it/sec, obj=4]
INFO - 13:28:00: ... 88%|████████▊ | 106/121 [00:00<00:00, 1564.00 it/sec, obj=5]
INFO - 13:28:00: ... 88%|████████▊ | 107/121 [00:00<00:00, 1564.64 it/sec, obj=6]
INFO - 13:28:00: ... 89%|████████▉ | 108/121 [00:00<00:00, 1565.35 it/sec, obj=7]
INFO - 13:28:00: ... 90%|█████████ | 109/121 [00:00<00:00, 1566.07 it/sec, obj=8]
INFO - 13:28:00: ... 91%|█████████ | 110/121 [00:00<00:00, 1566.17 it/sec, obj=9]
INFO - 13:28:00: ... 92%|█████████▏| 111/121 [00:00<00:00, 1566.90 it/sec, obj=0]
INFO - 13:28:00: ... 93%|█████████▎| 112/121 [00:00<00:00, 1567.33 it/sec, obj=1]
INFO - 13:28:00: ... 93%|█████████▎| 113/121 [00:00<00:00, 1567.84 it/sec, obj=2]
INFO - 13:28:00: ... 94%|█████████▍| 114/121 [00:00<00:00, 1568.65 it/sec, obj=3]
INFO - 13:28:00: ... 95%|█████████▌| 115/121 [00:00<00:00, 1569.39 it/sec, obj=4]
INFO - 13:28:00: ... 96%|█████████▌| 116/121 [00:00<00:00, 1569.36 it/sec, obj=5]
INFO - 13:28:00: ... 97%|█████████▋| 117/121 [00:00<00:00, 1569.97 it/sec, obj=6]
INFO - 13:28:00: ... 98%|█████████▊| 118/121 [00:00<00:00, 1570.62 it/sec, obj=7]
INFO - 13:28:00: ... 98%|█████████▊| 119/121 [00:00<00:00, 1570.90 it/sec, obj=8]
INFO - 13:28:00: ... 99%|█████████▉| 120/121 [00:00<00:00, 1571.53 it/sec, obj=9]
INFO - 13:28:00: ... 100%|██████████| 121/121 [00:00<00:00, 1572.20 it/sec, obj=10]
INFO - 13:28:00: Optimization result:
INFO - 13:28:00: Optimizer info:
INFO - 13:28:00: Status: None
INFO - 13:28:00: Message: None
INFO - 13:28:00: Number of calls to the objective function by the optimizer: 121
INFO - 13:28:00: Solution:
INFO - 13:28:00: Objective: -10.0
INFO - 13:28:00: Design space:
INFO - 13:28:00: +------+-------------+-------+-------------+---------+
INFO - 13:28:00: | name | lower_bound | value | upper_bound | type |
INFO - 13:28:00: +------+-------------+-------+-------------+---------+
INFO - 13:28:00: | x1 | -5 | -5 | 5 | integer |
INFO - 13:28:00: | x2 | -5 | -5 | 5 | integer |
INFO - 13:28:00: +------+-------------+-------+-------------+---------+
INFO - 13:28:00: *** End DOEScenario execution (time: 0:00:00.086613) ***
{'eval_jac': False, 'n_samples': 121, 'algo': 'fullfact'}
The optimum results can be found in the execution log. It is also possible to
extract them by invoking the Scenario.get_optimum()
method. It
returns a dictionary containing the optimum results for the
scenario under consideration:
opt_results = scenario.get_optimum()
print(
"The solution of P is (x*,f(x*)) = ({}, {})".format(
opt_results.x_opt, opt_results.f_opt
),
)
The solution of P is (x*,f(x*)) = ([-5. -5.], -10.0)
Available DOE algorithms¶
In order to get the list of available DOE algorithms, use:
algo_list = get_available_doe_algorithms()
print(f"Available algorithms: {algo_list}")
Available algorithms: ['CustomDOE', 'DiagonalDOE', 'OT_SOBOL', 'OT_RANDOM', 'OT_HASELGROVE', 'OT_REVERSE_HALTON', 'OT_HALTON', 'OT_FAURE', 'OT_MONTE_CARLO', 'OT_FACTORIAL', 'OT_COMPOSITE', 'OT_AXIAL', 'OT_OPT_LHS', 'OT_LHS', 'OT_LHSC', 'OT_FULLFACT', 'OT_SOBOL_INDICES', 'fullfact', 'ff2n', 'pbdesign', 'bbdesign', 'ccdesign', 'lhs']
Available post-processing¶
In order to get the list of available post-processing algorithms, use:
post_list = get_available_post_processings()
print(f"Available algorithms: {post_list}")
Available algorithms: ['BasicHistory', 'Compromise', 'ConstraintsHistory', 'Correlations', 'GradientSensitivity', 'HighTradeOff', 'MultiObjectiveDiagram', 'ObjConstrHist', 'OptHistoryView', 'ParallelCoordinates', 'ParetoFront', 'Petal', 'QuadApprox', 'Radar', 'RadarChart', 'Robustness', 'SOM', 'ScatterPareto', 'ScatterPlotMatrix', 'VariableInfluence']
You can also look at the examples:
Total running time of the script: ( 0 minutes 0.103 seconds)