Note
Click here to download the full example code
RBF regression¶
We want to approximate a discipline with two inputs and two outputs:
\(y_1=1+2x_1+3x_2\)
\(y_2=-1-2x_1-3x_2\)
over the unit hypercube \([0,1]\times[0,1]\).
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.mlearning.api import create_regression_model
from numpy import array
configure_logger()
Out:
<RootLogger root (INFO)>
Create the discipline to learn¶
We can implement this analytic discipline by means of the
AnalyticDiscipline
class.
expressions = {"y_1": "1+2*x_1+3*x_2", "y_2": "-1-2*x_1-3*x_2"}
discipline = create_discipline(
"AnalyticDiscipline", name="func", expressions=expressions
)
Create the input sampling space¶
We create the input sampling space by adding the variables one by one.
design_space = create_design_space()
design_space.add_variable("x_1", l_b=0.0, u_b=1.0)
design_space.add_variable("x_2", l_b=0.0, u_b=1.0)
Create the learning set¶
We can build a learning set by means of a
DOEScenario
with a full factorial design of
experiments. The number of samples can be equal to 9 for example.
scenario = create_scenario(
[discipline], "DisciplinaryOpt", "y_1", design_space, scenario_type="DOE"
)
scenario.execute({"algo": "fullfact", "n_samples": 9})
Out:
INFO - 07:18:56:
INFO - 07:18:56: *** Start DOEScenario execution ***
INFO - 07:18:56: DOEScenario
INFO - 07:18:56: Disciplines: func
INFO - 07:18:56: MDO formulation: DisciplinaryOpt
INFO - 07:18:56: Optimization problem:
INFO - 07:18:56: minimize y_1(x_1, x_2)
INFO - 07:18:56: with respect to x_1, x_2
INFO - 07:18:56: over the design space:
INFO - 07:18:56: +------+-------------+-------+-------------+-------+
INFO - 07:18:56: | name | lower_bound | value | upper_bound | type |
INFO - 07:18:56: +------+-------------+-------+-------------+-------+
INFO - 07:18:56: | x_1 | 0 | None | 1 | float |
INFO - 07:18:56: | x_2 | 0 | None | 1 | float |
INFO - 07:18:56: +------+-------------+-------+-------------+-------+
INFO - 07:18:56: Solving optimization problem with algorithm fullfact:
INFO - 07:18:56: Full factorial design required. Number of samples along each direction for a design vector of size 2 with 9 samples: 3
INFO - 07:18:56: Final number of samples for DOE = 9 vs 9 requested
INFO - 07:18:56: ... 0%| | 0/9 [00:00<?, ?it]
INFO - 07:18:56: ... 100%|██████████| 9/9 [00:00<00:00, 1429.23 it/sec, obj=6]
INFO - 07:18:56: Optimization result:
INFO - 07:18:56: Optimizer info:
INFO - 07:18:56: Status: None
INFO - 07:18:56: Message: None
INFO - 07:18:56: Number of calls to the objective function by the optimizer: 9
INFO - 07:18:56: Solution:
INFO - 07:18:56: Objective: 1.0
INFO - 07:18:56: Design space:
INFO - 07:18:56: +------+-------------+-------+-------------+-------+
INFO - 07:18:56: | name | lower_bound | value | upper_bound | type |
INFO - 07:18:56: +------+-------------+-------+-------------+-------+
INFO - 07:18:56: | x_1 | 0 | 0 | 1 | float |
INFO - 07:18:56: | x_2 | 0 | 0 | 1 | float |
INFO - 07:18:56: +------+-------------+-------+-------------+-------+
INFO - 07:18:56: *** End DOEScenario execution (time: 0:00:00.014820) ***
{'eval_jac': False, 'algo': 'fullfact', 'n_samples': 9}
Create the regression model¶
Then, we build the linear regression model from the database and displays this model.
dataset = scenario.export_to_dataset(opt_naming=False)
model = create_regression_model("RBFRegressor", data=dataset)
model.learn()
print(model)
Out:
RBFRegressor(epsilon=None, function='multiquadric', norm='euclidean', smooth=0.0)
based on the SciPy library
built from 9 learning samples
Predict output¶
Once it is built, we can use it for prediction.
input_value = {"x_1": array([1.0]), "x_2": array([2.0])}
output_value = model.predict(input_value)
print(output_value)
Out:
{'y_1': array([6.45029404])}
Total running time of the script: ( 0 minutes 0.037 seconds)