def generateIndividual(self):
individual = Individual()
individual.features = []
# HERE WE HANDLE THE SEARCH SPACE
# for i in range(self.n):
# individual.features.append(random.random())
react_temp = random.uniform(pbr_props.temperature_lower_bound,
pbr_props.temperature_upper_bound)
react_pres = random.uniform(pbr_props.pressure_lower_bound,
pbr_props.pressure_upper_bound)
react_buoh = random.uniform(pbr_props.mole_flows_lower_bound,
pbr_props.mole_flows_upper_bound)
react_acac = random.uniform(pbr_props.mole_flows_lower_bound,
pbr_props.mole_flows_upper_bound)
individual.features.append(react_temp)
individual.features.append(react_pres)
individual.features.append(react_buoh)
individual.features.append(react_acac)
individual.dominates = functools.partial(self.__dominates,
individual1=individual)
# self.calculate_objectives(individual)
return individual
def generateIndividual(self):
individual = Individual()
individual.features = []
for i in range(self.n):
individual.features.append(random.uniform(self.bounds[i][0], self.bounds[i][1]))
self.calculate_objectives(individual)
individual.dominates = functools.partial(self.__dominates, individual1=individual)
return individual
def generateIndividual(self):
individual = Individual()
individual.features = []
for i in range(30):
individual.features.append(random.random())
individual.dominates = functools.partial(self.__dominates, individual1=individual)
self.calculate_objectives(individual)
return individual
def generate_individual(self):
xy = CLUSTERING(self.all_data_matrix, self.individual_no)
centroids, label, no_of_cluster = xy.kmeans(
) #Generate each individual using k-means clustering
individual = Individual()
individual.no_of_Cluster = no_of_cluster
individual.labels = label
#individual.BHI = self.zdt_definitions.BHI(individual)
individual.features = []
for sub_list in centroids:
for feature in sub_list:
individual.features.append(feature)
print(individual.no_of_Cluster)
return individual
def generate_individual(self):
"""
Parameters
----------
Returns
-------
individual
a new random portfolio
"""
individual = Individual()
individual.features = self.init_population(self.portfolio_size, 1,
self.Q, self.Q_rfa,
self.C)[0]
return individual
def generateIndividual(self, new_msr, new_rv, new_bisize, nsol_K,
nsol_label, nsol):
individual = Individual()
individual.features = []
individual.ob1 = new_msr
individual.ob2 = new_rv
individual.ob3 = new_bisize
individual.K = nsol_K
individual.label = nsol_label
for i in range(len(nsol)):
individual.features.append(nsol[i])
individual.dominates = functools.partial(self.__dominates,
individual1=individual)
self.calculate_objectives(individual)
return individual
def generate_individual(self):
individual = Individual()
individual.id = ''.join(random.choice(string.ascii_uppercase + string.digits) for _ in range(10))
individual.features = [random.uniform(*x) for x in self.variables_range]
hyperparameters = ['criterion', 'max_depth', 'min_samples_split', 'max_leaf_nodes', 'class_weight']
individual.features = od(zip(hyperparameters, individual.features))
individual.features = decode(self.variables_range, **individual.features)
individual.creation_mode = "inicialization"
return individual
def generateIndividual(self):
xy = FCM(self.all_data_matrix, self.individual_no)
center, u, d, fpc, no_of_cluster, label = xy.FuzzyCMeans(
) #Generate each individual using Fuzzy C means
individual = Individual()
individual.partition_matrix = u
individual.distance_matrix = d
individual.no_of_Cluster = no_of_cluster
individual.fpc = fpc
individual.labels = label
#individual.BHI = self.zdt_definitions.BHI(individual)
individual.features = []
for sub_list in center:
for feature in sub_list:
individual.features.append(feature)
#print individual.no_of_Cluster , individual.features
individual.dominates = functools.partial(self.__dominates,
individual1=individual)
self.calculate_objectives(individual)
return individual
def generateIndividual(self):
individual = Individual()
individual.vardim = self.vardim
individual.bound = self.bound
individual.hos_opt = self.hos_opt
##-----------------初始化待优化的医院---------------##
'''初始化随机生成染色体（多个基因的list），长度为个体的维度'''
len = self.vardim
rnd = np.random.random(size=len) #n维0-1之间的向量
individual.features = np.zeros(len) #长尾为n的0维向量
#在上限、下限之间生成染色体的值
for i in xrange(0, len):
individual.features[i] = int(self.bound[0, i] + (self.bound[1, i] - self.bound[0, i]) * rnd[i])
#计算目标函数值
self.calculate_objectives(individual)
#进行支配非支配的计算
individual.dominates = functools.partial(self.__dominates, individual1=individual)
return individual
def generate_default_individual_entropy(self):
individual = Individual()
individual.id = ''.join(
random.choice(string.ascii_uppercase + string.digits)
for _ in range(10))
individual.features = [1, None, 2, None, None]
hyperparameters = [
'criterion', 'max_depth', 'min_samples_split', 'max_leaf_nodes',
'class_weight'
]
#individual.features = [1, None, 1, 0.00001, None]
#hyperparameters = ['criterion','max_depth', 'min_samples_leaf', 'min_impurity_decrease', 'class_weight']
individual.features = od(zip(hyperparameters, individual.features))
individual.features = decode(self.variables_range,
**individual.features)
individual.creation_mode = "inicialization"
return individual
def generate_individual(self):
individual = Individual()
individual.features = [random.uniform(*x) for x in self.variables_range]
return individual