linear_crossover

This function is a linear crossover between two individuals (represented by parent_0 and parent_1) in an optimization algorithm.

x_i_new, of_i_new, fit_i_new, neof, report = linear_crossover(of_function, parent_0, parent_1, n_dimensions, x_upper, x_lower, none_variable=None)

Input variables

Name	Description	Type
of_function	Objective function	Py function (def)
parent_0	First parent	List
parent_1	Second parent	List
n_dimensions	Crossover point	Integer
x_upper	Upper limit of the design variables	List
x_lower	Lower limit of the design variables	List
none_variable	None variable. Default is None. Use in objective function	None, List, Float, Dictionary, String or any

Output variables

Name	Description	Type
x_i_new	Update variables of the i agent	List
of_i_new	Update objective function value of the i agent	Float
fit_i_new	Update fitness value of the i agent	Float
neof	Number of evaluations of the objective function	Integer
report	Report about the crossover process	String

Example 1

# Import
from metapy_toolbox import linear_crossover

# Data
father1 = [1.2, 1.4, 2.5, 4.8, 2.4]
father2 = [2.3, 1.5, 3.5, 4.8, 2.4]
nDimensions = len(father1)
xUpper = [5, 5, 5, 5, 5]
xLower = [1, 1, 1, 1, 1]
noneVariable = None

def objFunction(x, _):
    """Example objective function"""
    x0 = x[0]
    x1 = x[1]
    of = x0 ** 2 + x1 ** 2
    return of

# Call function
xNew, ofNew, fitNew, neof, report = linear_crossover(objFunction, father1, father2, nDimensions, xUpper, xLower, noneVariable)

# Output details
print('x new ', xNew)
print('of new ', ofNew)
print('fit new', fitNew)
print('number of evalutions objective function', neof)

x new  [1.0, 1.0, 1.0, 1.0, 1.0]
of new  2.0
fit new 0.3333333333333333
number of evalutions objective function 3

To check the movement report just apply the following instruction.

# Report details
arq = "report_example.txt"

# Writing report
with open(arq, "w") as file:
    file.write(report)

Open report_example.txt.

    Crossover operator - Linear crossover
    current p0 = [1.2, 1.4, 2.5, 4.8, 2.4]
    current p1 = [2.3, 1.5, 3.5, 4.8, 2.4]
    Dimension 0: alpha_a = 0.6, beta_a = 1.15, neighbor_a = 1.75
    Dimension 0: alpha_b = 1.7999999999999998, beta_b = 1.15, neighbor_b = 0.6499999999999999
    Dimension 0: alpha_c = 0.6, beta_c = 3.4499999999999997, neighbor_c = 2.8499999999999996
    Dimension 1: alpha_a = 0.7, beta_a = 0.75, neighbor_a = 1.45
    Dimension 1: alpha_b = 2.0999999999999996, beta_b = 0.75, neighbor_b = 1.3499999999999996
    Dimension 1: alpha_c = 0.7, beta_c = 2.25, neighbor_c = 1.55
    Dimension 2: alpha_a = 1.25, beta_a = 1.75, neighbor_a = 3.0
    Dimension 2: alpha_b = 3.75, beta_b = 1.75, neighbor_b = 2.0
    Dimension 2: alpha_c = 1.25, beta_c = 5.25, neighbor_c = 4.0
    Dimension 3: alpha_a = 2.4, beta_a = 2.4, neighbor_a = 4.8
    Dimension 3: alpha_b = 7.199999999999999, beta_b = 2.4, neighbor_b = 4.799999999999999
    Dimension 3: alpha_c = 2.4, beta_c = 7.199999999999999, neighbor_c = 4.799999999999999
    Dimension 4: alpha_a = 1.2, beta_a = 1.2, neighbor_a = 2.4
    Dimension 4: alpha_b = 3.5999999999999996, beta_b = 1.2, neighbor_b = 2.3999999999999995
    Dimension 4: alpha_c = 1.2, beta_c = 3.5999999999999996, neighbor_c = 2.3999999999999995
    offspring a = [1.0, 1.0, 1.0, 1.0, 1.0], of_a 2.0
    offspring b = [1.0, 1.0, 1.0, 1.0, 1.0], of_b 2.0
    offspring c = [1.0, 1.0, 1.0, 1.0, 1.0], of_c 2.0
    update x = [1.0, 1.0, 1.0, 1.0, 1.0], of = 2.0, fit = 0.3333333333333333