# !/usr/bin/env python
# Created by "Thieu" at 12:24, 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 OriginalBBO(Optimizer):
"""
The original version of: Biogeography-Based Optimization (BBO)
Links:
1. https://ieeexplore.ieee.org/abstract/document/4475427
Hyper-parameters should fine tuned in approximate range to get faster convergen toward the global optimum:
+ p_m: [0.01, 0.2], Mutation probability
+ elites: [2, 5], Number of elites will be keep for next generation
Examples
~~~~~~~~
>>> import numpy as np
>>> from mealpy.bio_based.BBO import OriginalBBO
>>>
>>> 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
>>> p_m = 0.01
>>> elites = 2
>>> model = OriginalBBO(problem_dict1, epoch, pop_size, p_m, elites)
>>> best_position, best_fitness = model.solve()
>>> print(f"Solution: {best_position}, Fitness: {best_fitness}")
References
~~~~~~~~~~
[1] Simon, D., 2008. Biogeography-based optimization. IEEE transactions on evolutionary computation, 12(6), pp.702-713.
"""
def __init__(self, problem, epoch=10000, pop_size=100, p_m=0.01, elites=2, **kwargs):
"""
Initialize the algorithm components.
Args:
problem (dict): The problem dictionary
epoch (int): Maximum number of iterations, default = 10000
pop_size (int): Number of population size, default = 100
p_m (float): Mutation probability, default=0.01
elites (int): Number of elites will be keep for next generation, default=2
"""
super().__init__(problem, kwargs)
self.nfe_per_epoch = pop_size
self.sort_flag = False
self.epoch = epoch
self.pop_size = pop_size
self.p_m = p_m
self.elites = elites
self.mu = (self.pop_size + 1 - np.array(range(1, self.pop_size + 1))) / (self.pop_size + 1)
self.mr = 1 - self.mu
[docs] def evolve(self, epoch):
"""
The main operations (equations) of algorithm. Inherit from Optimizer class
Args:
epoch (int): The current iteration
"""
_, pop_elites, _ = self.get_special_solutions(self.pop, best=self.elites)
pop = []
for idx in range(0, self.pop_size):
# Probabilistic migration to the i-th position
pos_new = deepcopy(self.pop[idx][self.ID_POS])
for j in range(self.problem.n_dims):
if np.random.uniform() < self.mr[idx]: # Should we immigrate?
# Pick a position from which to emigrate (roulette wheel selection)
random_number = np.random.uniform() * np.sum(self.mu)
select = self.mu[0]
select_index = 0
while (random_number > select) and (select_index < self.pop_size - 1):
select_index += 1
select += self.mu[select_index]
# this is the migration step
pos_new[j] = self.pop[select_index][self.ID_POS][j]
noise = np.random.uniform(self.problem.lb, self.problem.ub)
pos_new = np.where(np.random.uniform(0, 1, self.problem.n_dims) < self.p_m, noise, pos_new)
pos_new = self.amend_position(pos_new)
pop.append([pos_new, None])
pop = self.update_fitness_population(pop)
# replace the solutions with their new migrated and mutated versions then Merge Populations
self.pop = self.get_sorted_strim_population(pop + pop_elites, self.pop_size)
[docs]class BaseBBO(OriginalBBO):
"""
My changed version of: Biogeography-Based Optimization (BBO)
Links:
1. https://ieeexplore.ieee.org/abstract/document/4475427
Hyper-parameters should fine tuned in approximate range to get faster convergen toward the global optimum:
+ p_m: [0.01, 0.2], Mutation probability
+ elites: [2, 5], Number of elites will be keep for next generation
Examples
~~~~~~~~
>>> import numpy as np
>>> from mealpy.bio_based.BBO import BaseBBO
>>>
>>> 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
>>> p_m = 0.01
>>> elites = 2
>>> model = BaseBBO(problem_dict1, epoch, pop_size, p_m, elites)
>>> best_position, best_fitness = model.solve()
>>> print(f"Solution: {best_position}, Fitness: {best_fitness}")
"""
def __init__(self, problem, epoch=10000, pop_size=100, p_m=0.01, elites=2, **kwargs):
"""
Initialize the algorithm components.
Args:
problem (dict): The problem dictionary
epoch (int): Maximum number of iterations, default = 10000
pop_size (int): Number of population size, default = 100
p_m (float): Mutation probability, default=0.01
elites (int): Number of elites will be keep for next generation, default=2
**kwargs ():
"""
super().__init__(problem, epoch, pop_size, p_m, elites, **kwargs)
[docs] def evolve(self, epoch):
"""
The main operations (equations) of algorithm. Inherit from Optimizer class
Args:
epoch (int): The current iteration
"""
_, pop_elites, _ = self.get_special_solutions(self.pop, best=self.elites)
list_fitness = [agent[self.ID_TAR][self.ID_FIT] for agent in self.pop]
pop = []
for idx in range(0, self.pop_size):
# Probabilistic migration to the i-th position
# Pick a position from which to emigrate (roulette wheel selection)
idx_selected = self.get_index_roulette_wheel_selection(list_fitness)
# this is the migration step
pos_new = np.where(np.random.uniform(0, 1, self.problem.n_dims) < self.mr[idx], self.pop[idx_selected][self.ID_POS], self.pop[idx][self.ID_POS])
# Mutation
temp = np.random.uniform(self.problem.lb, self.problem.ub)
pos_new = np.where(np.random.uniform(0, 1, self.problem.n_dims) < self.p_m, temp, pos_new)
pos_new = self.amend_position(pos_new)
pop.append([pos_new, None])
pop = self.update_fitness_population(pop)
# Replace the solutions with their new migrated and mutated versions then merge populations
self.pop = self.get_sorted_strim_population(pop + pop_elites, self.pop_size)