import numpy as np
from iOpt.trial import Point
from iOpt.trial import FunctionValue
from iOpt.trial import Trial
from iOpt.problem import Problem
import problems.Shekel4.shekel4_generation as shekelGen
[документация]class Shekel4(Problem):
"""
The Scheckel function is a multivariate, multimodal, continuous, deterministic function, given by the formula:
:math:`f(x) = \sum_{i=1}^{m}(c_{i}+\sum_{j=1}^{n}(x-a_{i})^{2})^{-1}`,
where :math:`m` – number of maxima of the function,
:math:`a, c` - parameters generated randomly.
In the generator, the dimensionality of the problem is 4.
"""
[документация] def __init__(self, function_number: int):
"""
Constructor of the Shekel problem class.
:param functionNumber: task number in the set, :math:`1 <= functionNumber <= 3`
"""
super(Shekel4, self).__init__()
self.name = "Shekel4"
self.dimension = 4
self.number_of_float_variables = self.dimension
self.number_of_discrete_variables = 0
self.number_of_objectives = 1
self.number_of_constraints = 0
self.fn = function_number
self.float_variable_names = np.ndarray(shape=(self.dimension,), dtype=str)
for i in range(self.dimension):
self.float_variable_names[i] = i
self.lower_bound_of_float_variables = np.ndarray(shape=(self.dimension,), dtype=np.double)
self.lower_bound_of_float_variables.fill(0)
self.upper_bound_of_float_variables = np.ndarray(shape=(self.dimension,), dtype=np.double)
self.upper_bound_of_float_variables.fill(10)
self.known_optimum = np.ndarray(shape=(1,), dtype=Trial)
pointfv = np.ndarray(shape=(self.dimension,), dtype=np.double)
pointfv.fill(4)
KOpoint = Point(pointfv, [])
KOfunV = np.ndarray(shape=(1,), dtype=FunctionValue)
KOfunV[0] = FunctionValue()
KOfunV[0] = self.calculate(KOpoint, KOfunV[0])
self.known_optimum[0] = Trial(KOpoint, KOfunV)
[документация] def calculate(self, point: Point, function_value: FunctionValue) -> FunctionValue:
"""
Calculating the value of the selected function at a given point
:param point: coordinates of the trial point where the value of the function will be calculated.
:param function_value: object defining the function number in the task and storing the function value.
:return: Calculated value of the function at point.
"""
res: np.double = 0
for i in range(shekelGen.maxI[self.fn - 1]):
den: np.double = 0
for j in range(self.dimension):
den = den + pow((point.float_variables[j] - shekelGen.a[i][j]), 2.0)
res = res - 1 / (den + shekelGen.c[i])
function_value.value = res
return function_value