Multi-layer perceptron

\[\begin{split}y = f_1(x_1,x_2,x_3,x_4,x_5) \ if \ x_0 == 1 \\ y = f_2(x_1,x_2,x_3,x_4,x_5,x_6) \ if \ x_0 == 2 \\ y = f_3(x_1,x_2,x_3,x_4,x_5,x_6,x_7) \ if \ x_0 == 3 \\\end{split}\]

\(x_0 \in \{1,2,3\} , x_1 \in [-5,-2], x_2 \in [-5,-1], x_3 \in [8,16,32,64,128,256], x_4 \in \{ReLU,SeLU,ISRLU\}, x_{5,6,7} \in [0,5] .\)

Usage

import matplotlib.pyplot as plt

from smt.applications.mixed_integer import MixedIntegerSamplingMethod
from smt.problems import HierarchicalNeuralNetwork
from smt.sampling_methods import LHS

problem = HierarchicalNeuralNetwork()
ds = problem.design_space
n_doe = 100
ds.seed = 42
samp = MixedIntegerSamplingMethod(
    LHS, ds, criterion="ese", random_state=ds.seed
)
xdoe = samp(n_doe)
x_corr, eval_is_acting = ds.correct_get_acting(xdoe)
y = problem(x=x_corr, kx=None, eval_is_acting=eval_is_acting)

plt.scatter(xdoe[:, 0], y)
plt.xlabel("x")
plt.ylabel("y")
plt.show()

Options

List of options

Option	Default	Acceptable values	Acceptable types	Description
ndim	1	None	[‘int’]
return_complex	False	None	[‘bool’]
name	HierarchicalNeuralNetwork	None	[‘str’]