Note
Click here to download the full example code
Save and Load¶
We want to build a regression model and save it on the disk.
This regression model is an approximation of the following 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 __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.mlearning.api import create_regression_model
from gemseo.mlearning.api import import_regression_model
from numpy import array
configure_logger()
<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})
INFO - 16:38:00:
INFO - 16:38:00: *** Start DOEScenario execution ***
INFO - 16:38:00: DOEScenario
INFO - 16:38:00: Disciplines: func
INFO - 16:38:00: MDO formulation: DisciplinaryOpt
INFO - 16:38:00: Optimization problem:
INFO - 16:38:00: minimize y_1(x_1, x_2)
INFO - 16:38:00: with respect to x_1, x_2
INFO - 16:38:00: over the design space:
INFO - 16:38:00: +------+-------------+-------+-------------+-------+
INFO - 16:38:00: | name | lower_bound | value | upper_bound | type |
INFO - 16:38:00: +------+-------------+-------+-------------+-------+
INFO - 16:38:00: | x_1 | 0 | None | 1 | float |
INFO - 16:38:00: | x_2 | 0 | None | 1 | float |
INFO - 16:38:00: +------+-------------+-------+-------------+-------+
INFO - 16:38:00: Solving optimization problem with algorithm fullfact:
INFO - 16:38:00: ... 0%| | 0/9 [00:00<?, ?it]
INFO - 16:38:00: ... 11%|█ | 1/9 [00:00<00:00, 325.59 it/sec, obj=1]
INFO - 16:38:00: ... 22%|██▏ | 2/9 [00:00<00:00, 533.02 it/sec, obj=2]
INFO - 16:38:00: ... 33%|███▎ | 3/9 [00:00<00:00, 688.80 it/sec, obj=3]
INFO - 16:38:00: ... 44%|████▍ | 4/9 [00:00<00:00, 805.24 it/sec, obj=2.5]
INFO - 16:38:00: ... 56%|█████▌ | 5/9 [00:00<00:00, 892.52 it/sec, obj=3.5]
INFO - 16:38:00: ... 67%|██████▋ | 6/9 [00:00<00:00, 967.88 it/sec, obj=4.5]
INFO - 16:38:00: ... 78%|███████▊ | 7/9 [00:00<00:00, 1030.94 it/sec, obj=4]
INFO - 16:38:00: ... 89%|████████▉ | 8/9 [00:00<00:00, 1084.50 it/sec, obj=5]
INFO - 16:38:00: ... 100%|██████████| 9/9 [00:00<00:00, 1129.96 it/sec, obj=6]
INFO - 16:38:00: Optimization result:
INFO - 16:38:00: Optimizer info:
INFO - 16:38:00: Status: None
INFO - 16:38:00: Message: None
INFO - 16:38:00: Number of calls to the objective function by the optimizer: 9
INFO - 16:38:00: Solution:
INFO - 16:38:00: Objective: 1.0
INFO - 16:38:00: Design space:
INFO - 16:38:00: +------+-------------+-------+-------------+-------+
INFO - 16:38:00: | name | lower_bound | value | upper_bound | type |
INFO - 16:38:00: +------+-------------+-------+-------------+-------+
INFO - 16:38:00: | x_1 | 0 | 0 | 1 | float |
INFO - 16:38:00: | x_2 | 0 | 0 | 1 | float |
INFO - 16:38:00: +------+-------------+-------+-------------+-------+
INFO - 16:38:00: *** End DOEScenario execution (time: 0:00:00.017737) ***
{'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()
/home/docs/checkouts/readthedocs.org/user_builds/gemseo/envs/stable/lib/python3.9/site-packages/gemseo/mlearning/transform/scaler/min_max_scaler.py:73: RuntimeWarning: divide by zero encountered in divide
self.coefficient = where(is_constant, nan_to_num(1 / l_b), 1 / delta)
Use it for prediction¶
input_value = {"x_1": array([1.0]), "x_2": array([2.0])}
print(model.predict(input_value))
{'y_1': array([6.45029404])}
Save the regression model¶
Lastly, we save the model.
directory = model.save()
print(directory)
r_b_f_regressor_DOEScenario
Load the regression model¶
In an other study, we could load this model.
loaded_model = import_regression_model(directory)
print(loaded_model)
RBFRegressor(epsilon=None, function=multiquadric, norm=euclidean, smooth=0.0)
based on the SciPy library
built from 0 learning samples
Use the loaded regression model¶
And use it!
print(loaded_model.predict(input_value))
{'y_1': array([6.45029404])}
Total running time of the script: ( 0 minutes 0.043 seconds)