# !/usr/bin/env python
# Created by "Thieu" at 14:14, 01/03/2021 ----------%
# Email: nguyenthieu2102@gmail.com %
# Github: https://github.com/thieu1995 %
# --------------------------------------------------%
import numpy as np
from mealpy.optimizer import Optimizer
[docs]class BaseACOR(Optimizer):
"""
The original version of: Ant Colony Optimization Continuous (ACOR)
Notes
~~~~~
+ Use Gaussian Distribution instead of random number (np.random.normal() function)
+ Amend solution when they went out of space
Hyper-parameters should fine tuned in approximate range to get faster convergen toward the global optimum:
+ sample_count (int): [pop_size/2, pop_size], Number of Newly Generated Samples, default = 50
+ inten_factor (float): [0.2, 1.0], Intensification Factor (Selection Pressure), (q in the paper), default = 0.5
+ zeta (int): [1, 2, 3], Deviation-Distance Ratio, default = 1
Examples
~~~~~~~~
>>> import numpy as np
>>> from mealpy.swarm_based.ACOR import BaseACOR
>>>
>>> def fitness_function(solution):
>>> return np.sum(solution**2)
>>>
>>> problem_dict1 = {
>>> "fit_func": fitness_function,
>>> "lb": [-10, -15, -4, -2, -8],
>>> "ub": [10, 15, 12, 8, 20],
>>> "minmax": "min",
>>> "verbose": True,
>>> }
>>>
>>> epoch = 1000
>>> pop_size = 50
>>> sample_count = 50
>>> inten_factor = 0.5
>>> zeta = 1.0
>>> model = BaseACOR(problem_dict1, epoch, pop_size, sample_count, inten_factor, zeta)
>>> best_position, best_fitness = model.solve()
>>> print(f"Solution: {best_position}, Fitness: {best_fitness}")
References
~~~~~~~~~~
[1] Socha, K. and Dorigo, M., 2008. Ant colony optimization for continuous domains.
European journal of operational research, 185(3), pp.1155-1173.
"""
def __init__(self, problem, epoch=10000, pop_size=100, sample_count=50, inten_factor=0.5, zeta=1.0, **kwargs):
"""
Args:
problem (dict): The problem dictionary
epoch (int): maximum number of iterations, default = 10000
pop_size (int): number of population size, default = 100
sample_count (int): Number of Newly Generated Samples, default = 50
inten_factor (float): Intensification Factor (Selection Pressure) (q in the paper), default = 0.5
zeta (float): Deviation-Distance Ratio, default = 1.0
"""
super().__init__(problem, kwargs)
self.nfe_per_epoch = pop_size
self.sort_flag = True
self.epoch = epoch
self.pop_size = pop_size
self.sample_count = sample_count
self.inten_factor = inten_factor
self.zeta = zeta
[docs] def evolve(self, epoch):
"""
The main operations (equations) of algorithm. Inherit from Optimizer class
Args:
epoch (int): The current iteration
"""
# Calculate Selection Probabilities
pop = self.pop[:self.pop_size]
pop_rank = np.array([i for i in range(1, self.pop_size + 1)])
qn = self.inten_factor * self.pop_size
matrix_w = 1 / (np.sqrt(2 * np.pi) * qn) * np.exp(-0.5 * ((pop_rank - 1) / qn) ** 2)
matrix_p = matrix_w / np.sum(matrix_w) # Normalize to find the probability.
# Means and Standard Deviations
matrix_pos = np.array([solution[self.ID_POS] for solution in pop])
matrix_sigma = []
for i in range(0, self.pop_size):
matrix_i = np.repeat(pop[i][self.ID_POS].reshape((1, -1)), self.pop_size, axis=0)
D = np.sum(np.abs(matrix_pos - matrix_i), axis=0)
temp = self.zeta * D / (self.pop_size - 1)
matrix_sigma.append(temp)
matrix_sigma = np.array(matrix_sigma)
# Generate Samples
pop_new = []
for idx in range(0, self.sample_count):
# Generate Samples
child = np.zeros(self.problem.n_dims)
for j in range(0, self.problem.n_dims):
idx = self.get_index_roulette_wheel_selection(matrix_p)
child[j] = pop[idx][self.ID_POS][j] + np.random.normal() * matrix_sigma[idx, j] # (1)
pos_new = self.amend_position(child) # (2)
pop_new.append([pos_new, None])
pop_new = self.update_fitness_population(pop_new)
self.pop = pop + pop_new