def createWidgets(self):
self.selectLabel = ttk.Label(self, text = "Select:")
self.selectLabel.grid(column = 0, row = 0, padx = 10, pady = 5)
self.select = Select(self, textvariable = self.selectVar)
self.select.grid(column = 1, row = 0, sticky = (E, W))
self.searchLabel = ttk.Label(self, text = "Search:")
self.searchLabel.grid(column = 2, row = 0, padx = 10, pady = 5)
self.search = Search(self, textvariable = self.searchVar)
self.search.grid(column = 3, row = 0, sticky = (E, W))
self.filestorage = FileStorage(self)
self.filetree = FileTree(self)
self.filetree.grid(column = 0, row = 1, rowspan = 4, sticky = (N, S, E, W), columnspan = 4)
self.scrollbar = ttk.Scrollbar(self, orient = VERTICAL, command = self.filetree.yview)
self.scrollbar.grid(column = 4, row = 1, rowspan = 4, sticky = (N, S, E))
self.filetree.configure(yscrollcommand = self.scrollbar.set)
self.tags = Tags(self)
self.tags.grid(column = 5, row = 2, sticky = (E, W), padx = 5)
self.tagsLab = ttk.Label(text = "Tags")
self.tagsLab.grid(column = 5, row = 1, padx = 5, pady = 2)
self.notes = Notes(self)
self.notes.grid(column = 5, row = 4, sticky = (N, S, E, W), padx = 5)
self.notesLab = ttk.Label(text = "Notes")
self.notesLab.grid(column = 5, row = 3, padx = 5, pady = 2)
self.scrollNotes = ttk.Scrollbar(self, orient = VERTICAL, command = self.notes.yview)
self.scrollNotes.grid(column = 6, row = 4, sticky = (N, S, W))
self.notes.configure(yscrollcommand = self.scrollNotes.set)
self.buttons = Buttons(self)
self.buttons.grid(row = 5, column = 0, columnspan = 5, pady = 5, sticky = (E, W))
self.statusBar = StatusBar(self)
self.statusBar.grid(row = 6, column = 0, columnspan = 5, padx = 5, pady = 5, sticky = (E, W))
def cul_single(self, person):
other1 = person.columns[1]
target_ron = PredModel.predFunc(person.drop(other1, axis=1), 1)
true_target = person.iloc[0, 0]
print('本样本真实ron损失：', true_target)
print('预测模型认为的ron损失', target_ron)
print(
'\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\n'
)
# 初始化随机种群
buildings = []
value_record = []
s_record = []
best_param = [i * 0 for i in range(34)]
percent_record = []
# 20为种群内个体数
for i in range(32):
p = random.uniform(0.01, 0.99)
temp = []
for j in range(19):
base_value = person.iloc[0, 2 + j]
right_value = define.limit[j][1] - person.iloc[0, 2 + j]
left_value = person.iloc[0, 2 + j] - define.limit[j][0]
if random.randint(0, 1):
value = base_value + p * right_value
else:
value = base_value - p * left_value
# value = random.uniform(define.limit[j][0], define.limit[j][1])
temp.append(value)
for j in range(19, 30):
temp.append(person.iloc[0, j + 2])
buildings.append(temp)
# 进化次数250次，建议不少于200次
a = time.time()
for i in range(1000):
# 计算每个个体的适应度函数的值
value, s_value = Fit(buildings, person).fitness()
temp_min = min(value)
value_record.append(temp_min)
index = value.index(temp_min)
temp_s = s_value[index]
s_record.append(temp_s)
reduce_percent = (target_ron - temp_min) / target_ron
percent_record.append(reduce_percent)
if reduce_percent > 0.301 and temp_s < 5:
# if temp_min <= min(value_record) and temp_s < 5:
# best_param = [buildings[index], temp_min, reduce_percent, temp_s, i]
best_param = buildings[index] + [
temp_min, reduce_percent, temp_s, i
]
break
print('本轮最小RON损失：', temp_min)
print('RON损失降幅:{:.1f}%'.format(reduce_percent * 100))
print('对应硫含量：', temp_s)
print('================round', i, 'finished==================\n')
# t用于存储最优个体的适应度，用以绘制最后的进化图像
# record用于存储每次进化最优的个体，方便最后打印最优基因型
# Select函数决定本代哪些个体可以活到下一代
buildings_new = Select(buildings, value, s_value).selection()
# 活下来的个体以0.7的概率进行交配产生子代，交配概率动态降低
children = Cross(buildings_new,
self.cross_p *
max(1 - reduce_percent / 0.4, 0.1)).crossover()
# 子代以0.3的基准概率基因突变，突变概率动态降低
buildings = Mutate(
children,
self.mutate_p * max(1 - reduce_percent / 0.4, 0.1)).mutation()
b = time.time()
print(b - a)
# print('\n最佳优化参数：', best_param[0], '\nRON损失：', best_param[1], '\n损失降幅：', best_param[2],
# '\n产品硫含量：', best_param[3], '\n所在轮次:', best_param[4])
print(best_param)
plt.title('RON损失值')
plt.grid(ls='--')
plt.xlabel("进化轮次", fontsize=14)
plt.ylabel("RON损失", fontsize=14)
plt.plot(value_record)
# plt.show()
plt.title('产品硫含量')
plt.grid(ls='--')
plt.xlabel("进化轮次", fontsize=14)
plt.ylabel("硫含量", fontsize=14)
plt.plot(s_record)
# plt.show()
plt.title('RON损失降低比率')
plt.grid(ls='--')
plt.xlabel("进化轮次", fontsize=14)
plt.ylabel("降损比例", fontsize=14)
plt.plot(percent_record)
# plt.show()
resultframe = pd.DataFrame([best_param])
resultframe.to_csv('./result/opt_result.csv',
mode='a',
index=False,
header=None)
prefer = input('请输入设计的传统性【0/1】：0为现代性强，1为传统性强： ')
t = []
record = []
# 进化次数250次，建议不少于200次
for i in range(250):
# 计算每个个体的适应度函数的值
value = Fit(buildings, rain, land, money, location, prefer).fitness()
temp_max = max(value)
# t用于存储最优个体的适应度，用以绘制最后的进化图像
t.append(temp_max)
index = value.index(temp_max)
# record用于存储每次进化最优的个体，方便最后打印最优基因型
record.append(buildings[index])
# Select函数决定本代哪些个体可以活到下一代
buildings_new = Select(buildings, value).selection()
# 活下来的个体以0.7的概率进行交配产生子代（交叉概率0.7）
children = Cross(buildings_new, 0.7).crossover()
# 子代以0.02的概率基因突变
buildings = Mutate(children, 0.02).mutation()
# 提取最优解的基因型
value_max = Fit(record, rain, land, money, location, prefer).fitness()
temp_max = max(value_max)
index = value_max.index(temp_max)
# 把最优解的基因型从数字编码翻译成题设给的字母，打印输出
result = translate(record[index])
print('最优解的基因型为:', result)
# 输出收敛过程及最大值的函数图像
plt.plot(t)
class GUI(Tk):
"represents GUI"
def __init__(self):
super().__init__()
self.option_add("*tearOff", FALSE)
self.initialized = False
self.title("Papyer")
x, y = 1500, 500
self.minsize(x, y)
placeWindow(self, x, y)
self.options = Options(self)
self["menu"] = TopMenu(self)
self.protocol("WM_DELETE_WINDOW", self.closeFun)
self.base = os.getcwd()
self.selectVar = StringVar()
self.searchVar = StringVar()
self.createWidgets()
self.columnconfigure(1, weight = 1)
self.columnconfigure(3, weight = 1)
self.columnconfigure(5, weight = 1)
self.rowconfigure(4, weight = 1)
self.bind("<Control-d>", lambda e: self.filetree.keepDuplicates())
self.bind("<Control-a>", lambda e: self.filetree.selectAll())
self.mainloop()
def refresh(self):
# do in a smarter way - check changes in the files
self.filestorage.save()
self.filetree.saveSettings()
selected = self.filetree.selection()
self.createWidgets()
self.filetree.selection_set(selected)
def createWidgets(self):
self.selectLabel = ttk.Label(self, text = "Select:")
self.selectLabel.grid(column = 0, row = 0, padx = 10, pady = 5)
self.select = Select(self, textvariable = self.selectVar)
self.select.grid(column = 1, row = 0, sticky = (E, W))
self.searchLabel = ttk.Label(self, text = "Search:")
self.searchLabel.grid(column = 2, row = 0, padx = 10, pady = 5)
self.search = Search(self, textvariable = self.searchVar)
self.search.grid(column = 3, row = 0, sticky = (E, W))
self.filestorage = FileStorage(self)
self.filetree = FileTree(self)
self.filetree.grid(column = 0, row = 1, rowspan = 4, sticky = (N, S, E, W), columnspan = 4)
self.scrollbar = ttk.Scrollbar(self, orient = VERTICAL, command = self.filetree.yview)
self.scrollbar.grid(column = 4, row = 1, rowspan = 4, sticky = (N, S, E))
self.filetree.configure(yscrollcommand = self.scrollbar.set)
self.tags = Tags(self)
self.tags.grid(column = 5, row = 2, sticky = (E, W), padx = 5)
self.tagsLab = ttk.Label(text = "Tags")
self.tagsLab.grid(column = 5, row = 1, padx = 5, pady = 2)
self.notes = Notes(self)
self.notes.grid(column = 5, row = 4, sticky = (N, S, E, W), padx = 5)
self.notesLab = ttk.Label(text = "Notes")
self.notesLab.grid(column = 5, row = 3, padx = 5, pady = 2)
self.scrollNotes = ttk.Scrollbar(self, orient = VERTICAL, command = self.notes.yview)
self.scrollNotes.grid(column = 6, row = 4, sticky = (N, S, W))
self.notes.configure(yscrollcommand = self.scrollNotes.set)
self.buttons = Buttons(self)
self.buttons.grid(row = 5, column = 0, columnspan = 5, pady = 5, sticky = (E, W))
self.statusBar = StatusBar(self)
self.statusBar.grid(row = 6, column = 0, columnspan = 5, padx = 5, pady = 5, sticky = (E, W))
def closeFun(self):
"ask for saving files on exit"
self.filetree.saveSettings()
self.filestorage.save()
self.options.save()
self.destroy()
bestsIndividuos = [[], []]
bestsIndividuosCrossing = [[], []]
individuoMutation = [[], []]
fits = []
fit1 = []
fit2 = []
porcent = 3
meta = 28
quantGenes = 36
quantIndividuos = 100
geracao = 300
contGeracao = 0
melhorFit1 = 0
melhorFit2 = 0
individuo = Population()
selecao = Select()
def convert_bin(listaBits):
copylistaBits = listaBits.copy()
copylistaBits.reverse()
new_list = []
count = 0
decimal = 0
newlist = []
base = 0
for bit in copylistaBits:
base = bit * 2
if base != 0:
decimal = decimal + base**count