#!/usr/bin/env python
# Created by "Thieu" at 10:06, 17/03/2020 ----------%
# Email: nguyenthieu2102@gmail.com %
# Github: https://github.com/thieu1995 %
# --------------------------------------------------%
import numpy as np
from mealpy.optimizer import Optimizer
[docs]class OriginalWOA(Optimizer):
"""
The original version of: Whale Optimization Algorithm (WOA)
Links:
1. https://doi.org/10.1016/j.advengsoft.2016.01.008
2. https://mathworks.com/matlabcentral/fileexchange/55667-the-whale-optimization-algorithm
Examples
~~~~~~~~
>>> import numpy as np
>>> from mealpy import FloatVar, WOA
>>>
>>> def objective_function(solution):
>>> return np.sum(solution**2)
>>>
>>> problem_dict = {
>>> "bounds": FloatVar(lb=(-10.,) * 30, ub=(10.,) * 30, name="delta"),
>>> "minmax": "min",
>>> "obj_func": objective_function
>>> }
>>>
>>> model = WOA.OriginalWOA(epoch=1000, pop_size=50)
>>> g_best = model.solve(problem_dict)
>>> print(f"Solution: {g_best.solution}, Fitness: {g_best.target.fitness}")
>>> print(f"Solution: {model.g_best.solution}, Fitness: {model.g_best.target.fitness}")
References
~~~~~~~~~~
[1] Mirjalili, S. and Lewis, A., 2016. The whale optimization algorithm. Advances in engineering software, 95, pp.51-67.
"""
def __init__(self, epoch: int = 10000, pop_size: int = 100, **kwargs: object) -> None:
"""
Args:
epoch (int): maximum number of iterations, default = 10000
pop_size (int): number of population size, default = 100
"""
super().__init__(**kwargs)
self.epoch = self.validator.check_int("epoch", epoch, [1, 100000])
self.pop_size = self.validator.check_int("pop_size", pop_size, [5, 10000])
self.set_parameters(["epoch", "pop_size"])
self.sort_flag = False
[docs] def evolve(self, epoch):
"""
The main operations (equations) of algorithm. Inherit from Optimizer class
Args:
epoch (int): The current iteration
"""
a = 2 - 2 * epoch / self.epoch # linearly decreased from 2 to 0
a2 = -1 - epoch / self.epoch
pop_new = []
for idx in range(0, self.pop_size):
r1 = self.generator.random(self.problem.n_dims)
r2 = self.generator.random(self.problem.n_dims)
A = 2 * a * r1 - a
C = 2 * r2
bb = 1
ll = (a2 - 1) * self.generator.random() + 1
pp = self.generator.random(self.problem.n_dims)
pos_new = self.pop[idx].solution.copy()
for jdx in range(0, self.problem.n_dims):
if pp[jdx] < 0.5:
if np.abs(A[jdx]) >= 1:
id_r = self.generator.choice(list(set(range(0, self.pop_size)) - {idx}))
D_X_rand = abs(C[jdx] * self.pop[id_r].solution[jdx] - self.pop[idx].solution[jdx])
pos_new[jdx] = self.pop[id_r].solution[jdx] - A[jdx] * D_X_rand
else:
D_Leader = abs(C[jdx] * self.g_best.solution[jdx] - self.pop[idx].solution[jdx])
pos_new[jdx] = self.g_best.solution[jdx] - A[jdx] * D_Leader
else:
D1 = abs(self.g_best.solution[jdx] - self.pop[idx].solution[jdx])
pos_new[jdx] = D1 * np.exp(bb * ll) * np.cos(ll * 2 * np.pi) + self.g_best.solution[jdx]
pos_new = self.correct_solution(pos_new)
agent = self.generate_empty_agent(pos_new)
pop_new.append(agent)
if self.mode not in self.AVAILABLE_MODES:
self.pop[idx].update(solution=pos_new, target=self.get_target(pos_new))
if self.mode in self.AVAILABLE_MODES:
self.pop = self.update_target_for_population(pop_new)
[docs]class DevWOA(Optimizer):
"""
The developed version of: Whale Optimization Algorithm (WOA)
Notes:
+ Hanlding simple vector instead of loop through whole dimensions
+ Using greedy to update position
Examples
~~~~~~~~
>>> import numpy as np
>>> from mealpy import FloatVar, WOA
>>>
>>> def objective_function(solution):
>>> return np.sum(solution**2)
>>>
>>> problem_dict = {
>>> "bounds": FloatVar(lb=(-10.,) * 30, ub=(10.,) * 30, name="delta"),
>>> "minmax": "min",
>>> "obj_func": objective_function
>>> }
>>>
>>> model = WOA.DevWOA(epoch=1000, pop_size=50)
>>> g_best = model.solve(problem_dict)
>>> print(f"Solution: {g_best.solution}, Fitness: {g_best.target.fitness}")
>>> print(f"Solution: {model.g_best.solution}, Fitness: {model.g_best.target.fitness}")
References
~~~~~~~~~~
[1] Mirjalili, S. and Lewis, A., 2016. The whale optimization algorithm. Advances in engineering software, 95, pp.51-67.
"""
def __init__(self, epoch: int = 10000, pop_size: int = 100, **kwargs: object) -> None:
"""
Args:
epoch (int): maximum number of iterations, default = 10000
pop_size (int): number of population size, default = 100
"""
super().__init__(**kwargs)
self.epoch = self.validator.check_int("epoch", epoch, [1, 100000])
self.pop_size = self.validator.check_int("pop_size", pop_size, [5, 10000])
self.set_parameters(["epoch", "pop_size"])
self.sort_flag = False
[docs] def evolve(self, epoch):
"""
The main operations (equations) of algorithm. Inherit from Optimizer class
Args:
epoch (int): The current iteration
"""
a = 2 - 2 * epoch / self.epoch # linearly decreased from 2 to 0
pop_new = []
for idx in range(0, self.pop_size):
r1 = self.generator.random(self.problem.n_dims)
r2 = self.generator.random(self.problem.n_dims)
A = 2 * a * r1 - a
C = 2 * r2
ll = self.generator.uniform(-1, 1)
bb = 1
# Get pos1
pos1 = self.g_best.solution - A * np.abs(C * self.g_best.solution - self.pop[idx].solution)
# Get pos2
id_r2 = self.generator.choice(list(set(range(0, self.pop_size)) - {idx}))
pos2 = self.pop[id_r2].solution - A * np.abs(C * self.pop[id_r2].solution - self.pop[idx].solution)
pos_new = np.where(np.abs(A) < 1, pos1, pos2)
# Get pos3
D1 = np.abs(self.g_best.solution - self.pop[idx].solution)
pos3 = self.g_best.solution + np.exp(bb * ll) * np.cos(2 * np.pi * ll) * D1
# Get final pos_new
pos_new = np.where(self.generator.random(size=self.problem.n_dims) < 0.5, pos_new, pos3)
# Correct solution
pos_new = self.correct_solution(pos_new)
agent = self.generate_empty_agent(pos_new)
pop_new.append(agent)
if self.mode not in self.AVAILABLE_MODES:
agent.target = self.get_target(pos_new)
self.pop[idx] = self.get_better_agent(agent, self.pop[idx], self.problem.minmax)
if self.mode in self.AVAILABLE_MODES:
pop_new = self.update_target_for_population(pop_new)
self.pop = self.greedy_selection_population(self.pop, pop_new, self.problem.minmax)
[docs]class HI_WOA(Optimizer):
"""
The original version of: Hybrid Improved Whale Optimization Algorithm (HI-WOA)
Links:
1. https://ieenp.explore.ieee.org/document/8900003
Hyper-parameters should fine-tune in approximate range to get faster convergence toward the global optimum:
+ feedback_max (int): maximum iterations of each feedback, default = 10
Examples
~~~~~~~~
>>> import numpy as np
>>> from mealpy import FloatVar, WOA
>>>
>>> def objective_function(solution):
>>> return np.sum(solution**2)
>>>
>>> problem_dict = {
>>> "bounds": FloatVar(lb=(-10.,) * 30, ub=(10.,) * 30, name="delta"),
>>> "minmax": "min",
>>> "obj_func": objective_function
>>> }
>>>
>>> model = WOA.HI_WOA(epoch=1000, pop_size=50, feedback_max = 10)
>>> g_best = model.solve(problem_dict)
>>> print(f"Solution: {g_best.solution}, Fitness: {g_best.target.fitness}")
>>> print(f"Solution: {model.g_best.solution}, Fitness: {model.g_best.target.fitness}")
References
~~~~~~~~~~
[1] Tang, C., Sun, W., Wu, W. and Xue, M., 2019, July. A hybrid improved whale optimization algorithm.
In 2019 IEEE 15th International Conference on Control and Automation (ICCA) (pp. 362-367). IEEE.
"""
def __init__(self, epoch: int = 10000, pop_size: int = 100, feedback_max: int = 10, **kwargs: object) -> None:
"""
Args:
epoch (int): maximum number of iterations, default = 10000
pop_size (int): number of population size, default = 100
feedback_max (int): maximum iterations of each feedback, default = 10
"""
super().__init__(**kwargs)
self.epoch = self.validator.check_int("epoch", epoch, [1, 100000])
self.pop_size = self.validator.check_int("pop_size", pop_size, [5, 10000])
self.feedback_max = self.validator.check_int("feedback_max", feedback_max, [2, 2+int(self.epoch/2)])
# The maximum of times g_best doesn't change -> need to change half of population
self.set_parameters(["epoch", "pop_size", "feedback_max"])
self.sort_flag = True
[docs] def initialize_variables(self):
self.n_changes = int(self.pop_size / 2)
self.dyn_feedback_count = 0
[docs] def evolve(self, epoch):
"""
The main operations (equations) of algorithm. Inherit from Optimizer class
Args:
epoch (int): The current iteration
"""
a = 2 + 2 * np.cos(np.pi / 2 * (1 + epoch / self.epoch)) # Eq. 8
pop_new = []
for idx in range(0, self.pop_size):
r1 = self.generator.random(self.problem.n_dims)
r2 = self.generator.random(self.problem.n_dims)
A = 2 * a * r1 - a
C = 2 * r2
ll = self.generator.uniform(-1, 1)
bb = 1
if self.generator.uniform() < 0.5:
# Get pos1
pos1 = self.g_best.solution - A * np.abs(C * self.g_best.solution - self.pop[idx].solution)
# Get pos2
id_r2 = self.generator.choice(list(set(range(0, self.pop_size)) - {idx}))
pos2 = self.pop[id_r2].solution - A * np.abs(C * self.pop[id_r2].solution - self.pop[idx].solution)
pos_new = np.where(np.abs(A) < 1, pos1, pos2)
else:
D1 = np.abs(self.g_best.solution - self.pop[idx].solution)
pos_new = self.g_best.solution + np.exp(bb * ll) * np.cos(2 * np.pi * ll) * D1
pos_new = self.correct_solution(pos_new)
agent = self.generate_empty_agent(pos_new)
pop_new.append(agent)
if self.mode not in self.AVAILABLE_MODES:
agent.target = self.get_target(pos_new)
self.pop[idx] = self.get_better_agent(agent, self.pop[idx], self.problem.minmax)
if self.mode in self.AVAILABLE_MODES:
pop_new = self.update_target_for_population(pop_new)
self.pop = self.greedy_selection_population(self.pop, pop_new, self.problem.minmax)
## Feedback Mechanism
current_best = self.get_best_agent(self.pop, self.problem.minmax)
if current_best.target.fitness == self.g_best.target.fitness:
self.dyn_feedback_count += 1
else:
self.dyn_feedback_count = 0
if self.dyn_feedback_count >= self.feedback_max:
idx_list = self.generator.choice(range(0, self.pop_size), self.n_changes, replace=False)
pop_child = self.generate_population(self.n_changes)
for idx_counter, idx in enumerate(idx_list):
self.pop[idx] = pop_child[idx_counter]
[docs]class OriginalWOAmM(Optimizer):
"""
The original version of: Whale Optimization Algorithm with Modified Mutualism (WOAmM)
References
~~~~~~~~~~
[1] Chakraborty, S., Saha, A. K., Sharma, S., Mirjalili, S., & Chakraborty, R. (2021).
A novel enhanced whale optimization algorithm for global optimization. Computers & Industrial Engineering, 153, 107086.
https://doi.org/10.1016/j.cie.2020.107086
"""
def __init__(self, epoch: int = 10000, pop_size: int = 100, mut_rand: bool = False, patience: int = 0,
restart_rate: float = 0.2, bound: str = "clip", **kwargs) -> None:
"""
Args:
epoch: Maximum number of iterations.
pop_size: Population size.
mut_rand: Whether mutualism random coefficients are generated per dimension.
patience: Number of stagnant epochs before restarting worst agents. Set 0 to disable.
restart_rate: Ratio of worst agents to restart when stagnation occurs.
bound: Boundary handling method. Supported: "clip", "reflect", "random".
"""
super().__init__(**kwargs)
self.epoch = self.validator.check_int("epoch", epoch, [1, 100000])
self.pop_size = self.validator.check_int("pop_size", pop_size, [5, 10000])
self.mut_rand = self.validator.check_bool("mut_rand", mut_rand)
self.patience = self.validator.check_int("patience", patience, [0, 100000])
self.restart_rate = self.validator.check_float("restart_rate", restart_rate, [0.0, 1.0])
self.bound = self.validator.check_str("bound", bound, ["clip", "reflect", "random"])
self.set_parameters(["epoch", "pop_size", "mut_rand", "patience", "restart_rate", "bound"])
self.is_parallelizable = False
self.sort_flag = False
[docs] def initialize_variables(self):
self._stall_count = 0
self._stall_fit = None
[docs] def before_main_loop(self):
if self.g_best is not None and self.g_best.target is not None:
self._stall_fit = self.g_best.target.fitness
[docs] def restart_population(self):
n_restart = int(round(self.pop_size * self.restart_rate))
n_restart = np.clip(n_restart, 0, self.pop_size - 1)
_, indices = self.get_sorted_population(self.pop, self.problem.minmax, return_index=True)
worst_ids = indices[-n_restart:]
new_agents = self.generate_population(n_restart)
for idx, agent in zip(worst_ids, new_agents):
self.pop[idx] = agent
[docs] def restart_on_stagnation(self):
if self.patience <= 0:
return
best = self.get_best_agent(self.pop, self.problem.minmax)
if self._stall_fit is None or self.compare_fitness(best.target.fitness, self._stall_fit, self.problem.minmax):
self._stall_fit = best.target.fitness
self._stall_count = 0
return
self._stall_count += 1
if self._stall_count >= self.patience:
self._stall_count = 0
self.restart_population()
best = self.get_best_agent(self.pop, self.problem.minmax)
self._stall_fit = best.target.fitness
[docs] def reflect_solution(self, solution: np.ndarray) -> np.ndarray:
lb = self.problem.lb
ub = self.problem.ub
x = solution.copy()
span = ub - lb
valid = span > 0
x[~valid] = lb[~valid]
if np.any(valid):
offset = (x[valid] - lb[valid]) % (2.0 * span[valid])
offset = np.where(offset > span[valid], 2.0 * span[valid] - offset, offset)
x[valid] = lb[valid] + offset
return x
[docs] def random_solution(self, solution: np.ndarray) -> np.ndarray:
x = solution.copy()
mask = (x < self.problem.lb) | (x > self.problem.ub)
if np.any(mask):
x[mask] = self.generator.uniform(self.problem.lb[mask], self.problem.ub[mask])
return x
[docs] def amend_solution(self, solution: np.ndarray) -> np.ndarray:
if self.bound == "clip":
return solution
if self.bound == "random":
return self.random_solution(solution)
return self.reflect_solution(solution)
[docs] def evolve(self, epoch):
"""
Execute one iteration of WOAmM: modified mutualism phase followed by standard WOA moves.
Args:
epoch (int): The current iteration
"""
# Modified mutualism phase
base_pop = self.pop
new_pop = [agent.copy() for agent in self.pop]
for idx in range(0, self.pop_size):
id_m, id_n = self.generator.choice(list(set(range(0, self.pop_size)) - {idx}), size=2, replace=False)
# Paper: X_bf = best among Xi, Xm, Xn
# X_other = worse of Xm, Xn (NOT Xi, since Xi is always updated via Eq.15)
three_agents = [(idx, base_pop[idx].target),
(id_m, base_pop[id_m].target),
(id_n, base_pop[id_n].target)]
# Find best among all three for X_bf
if self.problem.minmax == "min":
id_best = min(three_agents, key=lambda x: x[1].fitness)[0]
else:
id_best = max(three_agents, key=lambda x: x[1].fitness)[0]
# X_other is the worse of Xm, Xn (never Xi)
if self.compare_target(base_pop[id_m].target, base_pop[id_n].target, self.problem.minmax):
id_other = id_n # id_n is worse
else:
id_other = id_m # id_m is worse
pos_i = base_pop[idx].solution
pos_other = base_pop[id_other].solution
pos_best = base_pop[id_best].solution
mutual_vector = (pos_i + pos_other) / 2
bf1, bf2 = self.generator.integers(1, 3, 2)
if self.mut_rand:
rnd_i = self.generator.random(self.problem.n_dims)
rnd_j = self.generator.random(self.problem.n_dims)
else:
rnd_i = self.generator.random()
rnd_j = self.generator.random()
xi_new = pos_i + rnd_i * (pos_best - mutual_vector * bf1)
xj_new = pos_other + rnd_j * (pos_best - mutual_vector * bf2)
xi_new = self.correct_solution(xi_new)
xj_new = self.correct_solution(xj_new)
xi_target = self.get_target(xi_new)
xj_target = self.get_target(xj_new)
if self.compare_target(xi_target, new_pop[idx].target, self.problem.minmax):
new_pop[idx].update(solution=xi_new, target=xi_target)
if self.compare_target(xj_target, new_pop[id_other].target, self.problem.minmax):
new_pop[id_other].update(solution=xj_new, target=xj_target)
self.pop = new_pop
# Refresh global best before WOA movement (keep best-so-far)
current_best = self.get_best_agent(self.pop, self.problem.minmax)
self.g_best = self.get_better_agent(current_best, self.g_best, self.problem.minmax)
# Standard WOA phase (like OriginalWOA with vector A, C per dimension)
a = 2 - 2 * epoch / self.epoch # linearly decreased from 2 to 0
a2 = -1 - epoch / self.epoch
pop_new = []
for idx in range(0, self.pop_size):
r1 = self.generator.random(self.problem.n_dims)
r2 = self.generator.random(self.problem.n_dims)
A = 2 * a * r1 - a
C = 2 * r2
bb = 1
ll = (a2 - 1) * self.generator.random() + 1
pp = self.generator.random(self.problem.n_dims)
pos_new = self.pop[idx].solution.copy()
for jdx in range(0, self.problem.n_dims):
if pp[jdx] < 0.5:
if np.abs(A[jdx]) >= 1:
id_r = self.generator.choice(list(set(range(0, self.pop_size)) - {idx}))
D_X_rand = abs(C[jdx] * self.pop[id_r].solution[jdx] - self.pop[idx].solution[jdx])
pos_new[jdx] = self.pop[id_r].solution[jdx] - A[jdx] * D_X_rand
else:
D_Leader = abs(C[jdx] * self.g_best.solution[jdx] - self.pop[idx].solution[jdx])
pos_new[jdx] = self.g_best.solution[jdx] - A[jdx] * D_Leader
else:
D1 = abs(self.g_best.solution[jdx] - self.pop[idx].solution[jdx])
pos_new[jdx] = D1 * np.exp(bb * ll) * np.cos(ll * 2 * np.pi) + self.g_best.solution[jdx]
pos_new = self.correct_solution(pos_new)
agent = self.generate_empty_agent(pos_new)
pop_new.append(agent)
if self.mode not in self.AVAILABLE_MODES:
self.pop[idx].update(solution=pos_new, target=self.get_target(pos_new))
if self.mode in self.AVAILABLE_MODES:
self.pop = self.update_target_for_population(pop_new)
self.restart_on_stagnation()
[docs]class DevWOAmM(OriginalWOAmM):
"""
The developed version of: Whale Optimization Algorithm with Modified Mutualism (WOAmM)
This version replaces the population after the WOA phase (no greedy selection).
References
~~~~~~~~~~
[1] Chakraborty, S., Saha, A. K., Sharma, S., Mirjalili, S., & Chakraborty, R. (2021).
A novel enhanced whale optimization algorithm for global optimization. Computers & Industrial Engineering, 153, 107086.
https://doi.org/10.1016/j.cie.2020.107086
"""
def __init__(self, epoch: int = 10000, pop_size: int = 100, mut_rand: bool = False, patience: int = 0,
restart_rate: float = 0.2, bound: str = "clip", **kwargs) -> None:
"""
Args:
epoch: Maximum number of iterations.
pop_size: Population size.
mut_rand: Whether mutualism random coefficients are generated per dimension.
patience: Number of stagnant epochs before restarting worst agents. Set 0 to disable.
restart_rate: Ratio of worst agents to restart when stagnation occurs.
bound: Boundary handling method. Supported: "clip", "reflect", "random".
"""
super().__init__(epoch, pop_size, mut_rand, patience, restart_rate, bound, **kwargs)
[docs] def evolve(self, epoch):
"""
Execute one iteration of WOAmM: modified mutualism phase followed by standard WOA moves.
Args:
epoch (int): The current iteration
"""
# Modified mutualism phase
base_pop = self.pop
new_pop = [agent.copy() for agent in self.pop]
for idx in range(0, self.pop_size):
id_m, id_n = self.generator.choice(list(set(range(0, self.pop_size)) - {idx}), size=2, replace=False)
# Paper: X_bf = best among Xi, Xm, Xn
# X_other = worse of Xm, Xn (NOT Xi, since Xi is always updated via Eq.15)
three_agents = [(idx, base_pop[idx].target),
(id_m, base_pop[id_m].target),
(id_n, base_pop[id_n].target)]
# Find best among all three for X_bf
if self.problem.minmax == "min":
id_best = min(three_agents, key=lambda x: x[1].fitness)[0]
else:
id_best = max(three_agents, key=lambda x: x[1].fitness)[0]
# X_other is the worse of Xm, Xn (never Xi)
if self.compare_target(base_pop[id_m].target, base_pop[id_n].target, self.problem.minmax):
id_other = id_n # id_n is worse
else:
id_other = id_m # id_m is worse
pos_i = base_pop[idx].solution
pos_other = base_pop[id_other].solution
pos_best = base_pop[id_best].solution
mutual_vector = (pos_i + pos_other) / 2
bf1, bf2 = self.generator.integers(1, 3, 2)
if self.mut_rand:
rnd_i = self.generator.random(self.problem.n_dims)
rnd_j = self.generator.random(self.problem.n_dims)
else:
rnd_i = self.generator.random()
rnd_j = self.generator.random()
xi_new = pos_i + rnd_i * (pos_best - mutual_vector * bf1)
xj_new = pos_other + rnd_j * (pos_best - mutual_vector * bf2)
xi_new = self.correct_solution(xi_new)
xj_new = self.correct_solution(xj_new)
xi_target = self.get_target(xi_new)
xj_target = self.get_target(xj_new)
if self.compare_target(xi_target, new_pop[idx].target, self.problem.minmax):
new_pop[idx].update(solution=xi_new, target=xi_target)
if self.compare_target(xj_target, new_pop[id_other].target, self.problem.minmax):
new_pop[id_other].update(solution=xj_new, target=xj_target)
self.pop = new_pop
# Refresh global best before WOA movement (keep best-so-far)
current_best = self.get_best_agent(self.pop, self.problem.minmax)
self.g_best = self.get_better_agent(current_best, self.g_best, self.problem.minmax)
# Standard WOA phase (like OriginalWOA with vector A, C per dimension)
a = 2 - 2 * epoch / self.epoch # linearly decreased from 2 to 0
a2 = -1 - epoch / self.epoch
pop_new = []
for idx in range(0, self.pop_size):
r1 = self.generator.random(self.problem.n_dims)
r2 = self.generator.random(self.problem.n_dims)
A = 2 * a * r1 - a
C = 2 * r2
bb = 1
ll = (a2 - 1) * self.generator.random() + 1
pp = self.generator.random(self.problem.n_dims)
pos_new = self.pop[idx].solution.copy()
for jdx in range(0, self.problem.n_dims):
if pp[jdx] < 0.5:
if np.abs(A[jdx]) >= 1:
id_r = self.generator.choice(list(set(range(0, self.pop_size)) - {idx}))
D_X_rand = abs(C[jdx] * self.pop[id_r].solution[jdx] - self.pop[idx].solution[jdx])
pos_new[jdx] = self.pop[id_r].solution[jdx] - A[jdx] * D_X_rand
else:
D_Leader = abs(C[jdx] * self.g_best.solution[jdx] - self.pop[idx].solution[jdx])
pos_new[jdx] = self.g_best.solution[jdx] - A[jdx] * D_Leader
else:
D1 = abs(self.g_best.solution[jdx] - self.pop[idx].solution[jdx])
pos_new[jdx] = D1 * np.exp(bb * ll) * np.cos(ll * 2 * np.pi) + self.g_best.solution[jdx]
pos_new = self.correct_solution(pos_new)
agent = self.generate_empty_agent(pos_new)
pop_new.append(agent)
if self.mode not in self.AVAILABLE_MODES:
agent.target = self.get_target(pos_new)
self.pop[idx] = self.get_better_agent(agent, self.pop[idx], self.problem.minmax)
if self.mode in self.AVAILABLE_MODES:
pop_new = self.update_target_for_population(pop_new)
self.pop = self.greedy_selection_population(self.pop, pop_new, self.problem.minmax)
self.restart_on_stagnation()