# !/usr/bin/env python
# Created by "Thieu" at 11:16, 18/03/2020 ----------%
# Email: nguyenthieu2102@gmail.com %
# Github: https://github.com/thieu1995 %
# --------------------------------------------------%
import numpy as np
from copy import deepcopy
from mealpy.optimizer import Optimizer
[docs]class BaseSARO(Optimizer):
"""
My changed version of: Search And Rescue Optimization (SARO)
Notes
~~~~~
All third loop is removed
Hyper-parameters should fine tuned in approximate range to get faster convergen toward the global optimum:
+ se (float): [0.3, 0.8], social effect, default = 0.5
+ mu (int): [10, 100], maximum unsuccessful search number, default = 50
Examples
~~~~~~~~
>>> import numpy as np
>>> from mealpy.human_based.SARO import BaseSARO
>>>
>>> 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
>>> se = 0.5
>>> mu = 50
>>> model = BaseSARO(problem_dict1, epoch, pop_size, se, mu)
>>> best_position, best_fitness = model.solve()
>>> print(f"Solution: {best_position}, Fitness: {best_fitness}")
"""
def __init__(self, problem, epoch=10000, pop_size=100, se=0.5, mu=50, **kwargs):
"""
Args:
problem (dict): The problem dictionary
epoch (int): maximum number of iterations, default = 10000
pop_size (int): number of population size, default = 100
se (float): social effect, default = 0.5
mu (int): maximum unsuccessful search number, default = 50
"""
super().__init__(problem, kwargs)
self.nfe_per_epoch = 2 * pop_size
self.sort_flag = True
self.epoch = epoch
self.pop_size = pop_size
self.se = se
self.mu = mu
## Dynamic variable
self.dyn_USN = np.zeros(self.pop_size)
[docs] def initialization(self):
pop = self.create_population(pop_size=(2 * self.pop_size))
self.pop, self.g_best = self.get_global_best_solution(pop)
[docs] def amend_position(self, position=None):
"""
If solution out of bound at dimension x, then it will re-arrange to random location in the range of domain
Args:
position: vector position (location) of the solution.
Returns:
Amended position
"""
return np.where(np.logical_and(self.problem.lb <= position, position <= self.problem.ub),
position, np.random.uniform(self.problem.lb, self.problem.ub))
[docs] def evolve(self, epoch):
"""
The main operations (equations) of algorithm. Inherit from Optimizer class
Args:
epoch (int): The current iteration
"""
pop_x = deepcopy(self.pop[:self.pop_size])
pop_m = deepcopy(self.pop[self.pop_size:])
pop_new = []
for idx in range(self.pop_size):
## Social Phase
k = np.random.choice(list(set(range(0, 2 * self.pop_size)) - {idx}))
sd = pop_x[idx][self.ID_POS] - self.pop[k][self.ID_POS]
#### Remove third loop here, also using random flight back when out of bound
pos_new_1 = self.pop[k][self.ID_POS] + np.random.uniform() * sd
pos_new_2 = pop_x[idx][self.ID_POS] + np.random.uniform() * sd
pos_new = np.where(np.logical_and(np.random.uniform(0, 1, self.problem.n_dims) < self.se,
self.pop[k][self.ID_TAR] < pop_x[idx][self.ID_TAR]), pos_new_1, pos_new_2)
pos_new = self.amend_position(pos_new)
pop_new.append([pos_new, None])
pop_new = self.update_fitness_population(pop_new)
for idx in range(self.pop_size):
if self.compare_agent(pop_new[idx], pop_x[idx]):
pop_m[np.random.randint(0, self.pop_size)] = deepcopy(pop_x[idx])
pop_x[idx] = deepcopy(pop_new[idx])
self.dyn_USN[idx] = 0
else:
self.dyn_USN[idx] += 1
pop = deepcopy(pop_x) + deepcopy(pop_m)
pop_new = []
for idx in range(self.pop_size):
## Individual phase
k1, k2 = np.random.choice(list(set(range(0, 2 * self.pop_size)) - {idx}), 2, replace=False)
#### Remove third loop here, and flight back strategy now be a random
pos_new = self.g_best[self.ID_POS] + np.random.uniform() * (pop[k1][self.ID_POS] - pop[k2][self.ID_POS])
pos_new = self.amend_position(pos_new)
pop_new.append([pos_new, None])
pop_new = self.update_fitness_population(pop_new)
for idx in range(0, self.pop_size):
if self.compare_agent(pop_new[idx], pop_x[idx]):
pop_m[np.random.randint(0, self.pop_size)] = deepcopy(pop_x[idx])
pop_x[idx] = deepcopy(pop_new[idx])
self.dyn_USN[idx] = 0
else:
self.dyn_USN[idx] += 1
if self.dyn_USN[idx] > self.mu:
pop_x[idx] = self.create_solution()
self.dyn_USN[idx] = 0
self.pop = pop_x + pop_m
[docs]class OriginalSARO(BaseSARO):
"""
The original version of: Search And Rescue Optimization (SARO)
Links:
1. https://doi.org/10.1155/2019/2482543
Hyper-parameters should fine tuned in approximate range to get faster convergen toward the global optimum:
+ se (float): [0.3, 0.8], social effect, default = 0.5
+ mu (int): [10, 100], maximum unsuccessful search number, default = 50
Examples
~~~~~~~~
>>> import numpy as np
>>> from mealpy.human_based.SARO import OriginalSARO
>>>
>>> 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
>>> se = 0.5
>>> mu = 50
>>> model = OriginalSARO(problem_dict1, epoch, pop_size, se, mu)
>>> best_position, best_fitness = model.solve()
>>> print(f"Solution: {best_position}, Fitness: {best_fitness}")
References
~~~~~~~~~~
[1] Shabani, A., Asgarian, B., Gharebaghi, S.A., Salido, M.A. and Giret, A., 2019. A new optimization
algorithm based on search and rescue operations. Mathematical Problems in Engineering, 2019.
"""
def __init__(self, problem, epoch=10000, pop_size=100, se=0.5, mu=50, **kwargs):
"""
Args:
problem (dict): The problem dictionary
epoch (int): maximum number of iterations, default = 10000
pop_size (int): number of population size, default = 100
se (float): social effect, default = 0.5
mu (int): maximum unsuccessful search number, default = 50
"""
super().__init__(problem, epoch, pop_size, se, mu, **kwargs)
[docs] def evolve(self, epoch):
"""
The main operations (equations) of algorithm. Inherit from Optimizer class
Args:
epoch (int): The current iteration
"""
pop_x = deepcopy(self.pop[:self.pop_size])
pop_m = deepcopy(self.pop[self.pop_size:])
pop_new = []
for idx in range(self.pop_size):
## Social Phase
k = np.random.choice(list(set(range(0, 2 * self.pop_size)) - {idx}))
sd = pop_x[idx][self.ID_POS] - self.pop[k][self.ID_POS]
j_rand = np.random.randint(0, self.problem.n_dims)
r1 = np.random.uniform(-1, 1)
pos_new = deepcopy(pop_x[idx][self.ID_POS])
for j in range(0, self.problem.n_dims):
if np.random.uniform() < self.se or j == j_rand:
if self.compare_agent(self.pop[k], pop_x[idx]):
pos_new[j] = self.pop[k][self.ID_POS][j] + r1 * sd[j]
else:
pos_new[j] = pop_x[idx][self.ID_POS][j] + r1 * sd[j]
if pos_new[j] < self.problem.lb[j]:
pos_new[j] = (pop_x[idx][self.ID_POS][j] + self.problem.lb[j]) / 2
if pos_new[j] > self.problem.ub[j]:
pos_new[j] = (pop_x[idx][self.ID_POS][j] + self.problem.ub[j]) / 2
pos_new = self.amend_position(pos_new)
pop_new.append([pos_new, None])
pop_new = self.update_fitness_population(pop_new)
for idx in range(0, self.pop_size):
if self.compare_agent(pop_new[idx], pop_x[idx]):
pop_m[np.random.randint(0, self.pop_size)] = deepcopy(pop_x[idx])
pop_x[idx] = deepcopy(pop_new[idx])
self.dyn_USN[idx] = 0
else:
self.dyn_USN[idx] += 1
## Individual phase
pop = deepcopy(pop_x) + deepcopy(pop_m)
pop_new = []
for idx in range(0, self.pop_size):
k, m = np.random.choice(list(set(range(0, 2 * self.pop_size)) - {idx}), 2, replace=False)
pos_new = pop_x[idx][self.ID_POS] + np.random.uniform() * (pop[k][self.ID_POS] - pop[m][self.ID_POS])
for j in range(0, self.problem.n_dims):
if pos_new[j] < self.problem.lb[j]:
pos_new[j] = (pop_x[idx][self.ID_POS][j] + self.problem.lb[j]) / 2
if pos_new[j] > self.problem.ub[j]:
pos_new[j] = (pop_x[idx][self.ID_POS][j] + self.problem.ub[j]) / 2
pos_new = self.amend_position(pos_new)
pop_new.append([pos_new, None])
pop_new = self.update_fitness_population(pop_new)
for idx in range(0, self.pop_size):
if self.compare_agent(pop_new[idx], pop_x[idx]):
pop_m[np.random.randint(0, self.pop_size)] = pop_x[idx]
pop_x[idx] = deepcopy(pop_new[idx])
self.dyn_USN[idx] = 0
else:
self.dyn_USN[idx] += 1
if self.dyn_USN[idx] > self.mu:
pop_x[idx] = self.create_solution()
self.dyn_USN[idx] = 0
self.pop = pop_x + pop_m