# Random forest 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 $$[0,1]\times[0,1]$$.

## Import¶

from __future__ import annotations

from numpy import array

from gemseo import configure_logger
from gemseo import create_design_space
from gemseo import create_discipline
from gemseo import create_scenario
from gemseo.mlearning import create_regression_model

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()


## 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 - 09:04:46:
INFO - 09:04:46: *** Start DOEScenario execution ***
INFO - 09:04:46: DOEScenario
INFO - 09:04:46:    Disciplines: func
INFO - 09:04:46:    MDO formulation: DisciplinaryOpt
INFO - 09:04:46: Optimization problem:
INFO - 09:04:46:    minimize y_1(x_1, x_2)
INFO - 09:04:46:    with respect to x_1, x_2
INFO - 09:04:46:    over the design space:
INFO - 09:04:46:       +------+-------------+-------+-------------+-------+
INFO - 09:04:46:       | Name | Lower bound | Value | Upper bound | Type  |
INFO - 09:04:46:       +------+-------------+-------+-------------+-------+
INFO - 09:04:46:       | x_1  |      0      |  None |      1      | float |
INFO - 09:04:46:       | x_2  |      0      |  None |      1      | float |
INFO - 09:04:46:       +------+-------------+-------+-------------+-------+
INFO - 09:04:46: Solving optimization problem with algorithm fullfact:
INFO - 09:04:46:     11%|█         | 1/9 [00:00<00:00, 363.46 it/sec, obj=1]
INFO - 09:04:46:     22%|██▏       | 2/9 [00:00<00:00, 578.13 it/sec, obj=2]
INFO - 09:04:46:     33%|███▎      | 3/9 [00:00<00:00, 732.54 it/sec, obj=3]
INFO - 09:04:46:     44%|████▍     | 4/9 [00:00<00:00, 840.08 it/sec, obj=2.5]
INFO - 09:04:46:     56%|█████▌    | 5/9 [00:00<00:00, 924.26 it/sec, obj=3.5]
INFO - 09:04:46:     67%|██████▋   | 6/9 [00:00<00:00, 994.62 it/sec, obj=4.5]
INFO - 09:04:46:     78%|███████▊  | 7/9 [00:00<00:00, 1051.77 it/sec, obj=4]
INFO - 09:04:46:     89%|████████▉ | 8/9 [00:00<00:00, 1095.23 it/sec, obj=5]
INFO - 09:04:46:    100%|██████████| 9/9 [00:00<00:00, 1135.33 it/sec, obj=6]
INFO - 09:04:46: Optimization result:
INFO - 09:04:46:    Optimizer info:
INFO - 09:04:46:       Status: None
INFO - 09:04:46:       Message: None
INFO - 09:04:46:       Number of calls to the objective function by the optimizer: 9
INFO - 09:04:46:    Solution:
INFO - 09:04:46:       Objective: 1.0
INFO - 09:04:46:       Design space:
INFO - 09:04:46:          +------+-------------+-------+-------------+-------+
INFO - 09:04:46:          | Name | Lower bound | Value | Upper bound | Type  |
INFO - 09:04:46:          +------+-------------+-------+-------------+-------+
INFO - 09:04:46:          | x_1  |      0      |   0   |      1      | float |
INFO - 09:04:46:          | x_2  |      0      |   0   |      1      | float |
INFO - 09:04:46:          +------+-------------+-------+-------------+-------+
INFO - 09:04:46: *** End DOEScenario execution (time: 0:00:00.019556) ***

{'eval_jac': False, 'n_samples': 9, 'algo': 'fullfact'}


## Create the regression model¶

Then, we build the linear regression model from the database and displays this model.

dataset = scenario.to_dataset(opt_naming=False)
model = create_regression_model("RandomForestRegressor", data=dataset)
model.learn()
model

RandomForestRegressor(n_estimators=100, random_state=0)
• based on the scikit-learn 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)
output_value

{'y_1': array([5.585])}


Total running time of the script: (0 minutes 0.166 seconds)

Gallery generated by Sphinx-Gallery