def update(self, space: Space, function: Function, iteration: int,
n_iterations: int) -> None:
"""Wraps Henry Gas Solubility Optimization over all agents and variables.
Args:
space: Space containing agents and update-related information.
function: A Function object that will be used as the objective function.
iteration: Current iteration.
n_iterations: Maximum number of iterations.
"""
# Creates n-wise clusters
clusters = g.n_wise(space.agents, self.pressure.shape[1])
# Iterates through all clusters
for i, cluster in enumerate(clusters):
# Calculates the system's current temperature (eq. 8)
T = np.exp(-iteration / n_iterations)
# Updates Henry's coefficient (eq. 8)
self.coefficient[i] *= np.exp(-self.constant[i] *
(1 / T - 1 / 298.15))
# Transforms the cluster into a list and sorts it
cluster = list(cluster)
cluster.sort(key=lambda x: x.fit)
# Iterates through all agents in cluster
for j, agent in enumerate(cluster):
# Calculates agent's solubility (eq. 9)
solubility = self.K * self.coefficient[i] * self.pressure[i][j]
# Updates agent's position (eq. 10)
agent.position = self._update_position(agent, cluster[0],
space.best_agent,
solubility)
# Clips agent's limits
agent.clip_by_bound()
# Re-calculates its fitness
agent.fit = function(agent.position)
# Re-sorts the whole space
space.agents.sort(key=lambda x: x.fit)
# Generates a uniform random number
r1 = r.generate_uniform_random_number()
# Calculates the number of worst agents (eq. 11)
N = int(len(space.agents) * (r1 * (0.2 - 0.1) + 0.1))
# Iterates through every bad agent
for agent in space.agents[-N:]:
# Generates another uniform random number
r2 = r.generate_uniform_random_number()
# Updates bad agent's position (eq. 12)
agent.position = agent.lb + r2 * (agent.ub - agent.lb)
def test_n_wise():
list = [1, 2, 3, 4]
pairs = general.n_wise(list)
for _ in pairs:
pass
assert type(pairs).__name__ == 'callable_iterator' or 'generator'
def _update(self, agents, function):
"""Method that wraps selection, crossover and mutation over all agents and variables.
Args:
agents (list): List of agents.
function (Function): A Function object that will be used as the objective function.
"""
# Creating a list to hold the new population
new_agents = []
# Retrieving the number of agents
n_agents = len(agents)
# Calculates a list of fitness from every agent
fitness = [agent.fit + c.EPSILON for agent in agents]
# Selects the parents
selected = self._roulette_selection(n_agents, fitness)
# For every pair of selected parents
for s in g.n_wise(selected):
# Performs the crossover
alpha, beta = self._crossover(agents[s[0]], agents[s[1]])
# Performs the mutation
alpha, beta = self._mutation(alpha, beta)
# Checking `alpha` limits
alpha.clip_limits()
# Checking `beta` limits
beta.clip_limits()
# Calculates new fitness for `alpha`
alpha.fit = function(alpha.position)
# Calculates new fitness for `beta`
beta.fit = function(beta.position)
# Appends the mutated agents to the children
new_agents.extend([alpha, beta])
# Joins both populations
agents += new_agents
# Sorting agents
agents.sort(key=lambda x: x.fit)
return agents[:n_agents]
def _crossover(self, space):
"""Crossover a number of individuals pre-selected through a tournament procedure (p. 101).
Args:
space (TreeSpace): A TreeSpace object.
agents (list): Current iteration agents.
trees (list): Current iteration trees.
"""
# Calculates a list of current trees' fitness
fitness = [agent.fit for agent in space.agents]
# Number of individuals to be crossovered
n_individuals = int(space.n_trees * self.p_crossover)
# Checks if `n_individuals` is an odd number
if n_individuals % 2 != 0:
# If it is, increase it by one
n_individuals += 1
# Gathers a list of selected individuals to be replaced
selected = g.tournament_selection(fitness, n_individuals)
# For every pair in selected individuals
for s in g.n_wise(selected):
# Calculates the amount of father nodes
father_nodes = space.trees[s[0]].n_nodes
# Calculate the amount of mother nodes
mother_nodes = space.trees[s[1]].n_nodes
# Checks if both trees have more than one node
if (father_nodes > 1) and (mother_nodes > 1):
# Prunning father nodes
max_f_nodes = self._prune_nodes(father_nodes)
# Prunning mother nodes
max_m_nodes = self._prune_nodes(mother_nodes)
# Apply the crossover operation
space.trees[s[0]], space.trees[s[1]] = self._cross(
space.trees[s[0]], space.trees[s[1]], max_f_nodes,
max_m_nodes)
def update(self, space: Space, function: Function) -> None:
"""Wraps Genetic Algorithm over all agents and variables.
Args:
space: Space containing agents and update-related information.
function: A Function object that will be used as the objective function.
"""
# Creates a list to hold the new population
new_agents = []
# Retrieves the number of agents
n_agents = len(space.agents)
# Calculates a list of fitness from every agent
fitness = [agent.fit + c.EPSILON for agent in space.agents]
# Selects the parents
selected = self._roulette_selection(n_agents, fitness)
# For every pair of selected parents
for s in g.n_wise(selected):
# Performs the crossover and mutation
alpha, beta = self._crossover(space.agents[s[0]],
space.agents[s[1]])
alpha, beta = self._mutation(alpha, beta)
# Checking `alpha` and `beta` limits
alpha.clip_by_bound()
beta.clip_by_bound()
# Calculates new fitness for `alpha` and `beta`
alpha.fit = function(alpha.position)
beta.fit = function(beta.position)
# Appends the mutated agents to the children
new_agents.extend([alpha, beta])
# Joins both populations, sort agents and gathers best `n_agents`
space.agents += new_agents
space.agents.sort(key=lambda x: x.fit)
space.agents = space.agents[:n_agents]
def _update(self, agents, best_agent, function, coefficient, pressure,
constant, iteration, n_iterations):
"""Method that wraps Henry Gas Solubility Optimization over all agents and variables.
Args:
agents (list): List of agents.
best_agent (Agent): Global best agent.
function (Function): A Function object that will be used as the objective function.
coefficient (np.array): Henry's coefficient array.
pressure (np.array): Partial pressure array.
constant (np.array): Constants array.
iteration (int): Current iteration.
n_iterations (int): Maximum number of iterations.
"""
# Creates n-wise clusters
clusters = g.n_wise(agents, pressure.shape[1])
# Iterates through all clusters
for i, cluster in enumerate(clusters):
# Calculates the system's current temperature (eq. 8)
T = np.exp(-iteration / n_iterations)
# Updates Henry's coefficient (eq. 8)
coefficient[i] *= np.exp(-constant[i] * (1 / T - 1 / 298.15))
# Transforms the cluster into a list and sorts it
cluster = list(cluster)
cluster.sort(key=lambda x: x.fit)
# Iterates through all agents in cluster
for j, agent in enumerate(cluster):
# Calculates agent's solubility (eq. 9)
solubility = self.K * coefficient[i] * pressure[i][j]
# Updates agent's position (eq. 10)
agent.position = self._update_position(agent, cluster[0],
best_agent, solubility)
# Clips agent's limits
agent.clip_limits()
# Re-calculates its fitness
agent.fit = function(agent.position)
# Re-sorts the whole space
agents.sort(key=lambda x: x.fit)
# Generates a uniform random number
r1 = r.generate_uniform_random_number()
# Calculates the number of worst agents (eq. 11)
N = int(len(agents) * (r1 * (0.2 - 0.1) + 0.1))
# Iterates through every bad agent
for agent in agents[-N:]:
# Generates another uniform random number
r2 = r.generate_uniform_random_number()
# Updates bad agent's position (eq. 12)
agent.position = agent.lb + r2 * (agent.ub - agent.lb)
import opytimizer.math.general as g
# Creates a list for pairwising
individuals = [1, 2, 3, 4]
# Creates pairwise from list
for pair in g.n_wise(individuals, 2):
# Outputting pairs
print(f"Pair: {pair}")
# Performs a tournmanet selection over list
selected = g.tournament_selection(individuals, 2)
# Outputting selected individuals
print(f"Selected: {selected}")