class MyUI(QtWidgets.QMainWindow, Ui_MainWindow):
def __init__(self):
super(MyUI, self).__init__()
self.setupUi(self)
self.figure = plt.figure() #可选参数,facecolor为背景颜色
self.canvas = FigureCanvas(self.figure)
self.timer = QTimer(self)
self.chooseMazeButton = QtWidgets.QPushButton()
self.chooseMazeButton.setObjectName("chooseMazeButton")
self.chooseMazeButton.setText("选择迷宫")
self.chooseAgentButton = QtWidgets.QPushButton()
self.chooseAgentButton.setObjectName("chooseAgentButton")
self.chooseAgentButton.setText("选择智能体")
self.displayMazeButton = QtWidgets.QPushButton()
self.displayMazeButton.setObjectName("displayMazeButton")
self.displayMazeButton.setText("显示迷宫")
self.displayResultButton = QtWidgets.QPushButton()
self.displayResultButton.setObjectName("displayResultButton")
self.displayResultButton.setText("显示结果")
self.mazeFileName = QtWidgets.QLabel()
self.mazeFileName.setObjectName("mazeFileName")
self.mazeFileName.setText("当前迷宫文件：无")
self.agentFileName = QtWidgets.QLabel()
self.agentFileName.setObjectName("agentFileName")
self.agentFileName.setText("当前智能体文件：无")
self.horizontalLayout.addWidget(self.canvas)
self.verticalLayout = QtWidgets.QVBoxLayout()
self.verticalLayout.setObjectName("verticalLayout")
self.horizontalLayout.addLayout(self.verticalLayout)
self.verticalLayout.addWidget(self.chooseMazeButton)
self.verticalLayout.addWidget(self.mazeFileName)
self.verticalLayout.addWidget(self.chooseAgentButton)
self.verticalLayout.addWidget(self.agentFileName)
self.verticalLayout.addWidget(self.displayMazeButton)
self.verticalLayout.addWidget(self.displayResultButton)
self.chooseMazeButton.clicked.connect(self.ChooseMaze)
self.chooseAgentButton.clicked.connect(self.ChooseAgent)
self.displayMazeButton.clicked.connect(self.DisplayMaze)
self.displayResultButton.clicked.connect(self.DisplayResult)
self.timer.timeout.connect(self.DisplayNext)
def ChooseMaze(self):
fileName = QFileDialog.getOpenFileName(self, "选取文件", './',
"All Files (*)")
fileName = fileName[0]
self.mazeFile = fileName
self.maze = np.load(fileName)
self.mazeFileName.setText("当前迷宫文件：" + fileName.split('/')[-1])
def ChooseAgent(self):
fileName = QFileDialog.getOpenFileName(self, "选取文件", './',
"All Files (*)")
fileName = fileName[0]
self.qTable = np.load(fileName)
self.agentFileName.setText("当前智能体文件：" + fileName.split('/')[-1])
def DisplayMaze(self):
if self.mazeFileName.text()[-1] == '无':
QMessageBox.warning(self, "Maze", "请先选择迷宫文件！", QMessageBox.Ok)
else:
self.maze = np.load(self.mazeFile)
self.SetCanvas(self.maze)
def DisplayResult(self):
if self.mazeFileName.text()[-1] == '无':
QMessageBox.warning(self, "Maze", "请先选择迷宫文件！", QMessageBox.Ok)
elif self.agentFileName.text()[-1] == '无':
QMessageBox.warning(self, "Maze", "请先选择智能体文件！", QMessageBox.Ok)
else:
self.agent = Agent(self.maze)
self.agent.maze.SetRewardAndNextStateDict()
self.agent.qTable = self.qTable
state = 0
self.stateList = []
while self.agent.maze.GetState(state) != 0.9:
self.stateList.append(state)
action = np.argmax(self.agent.qTable[:, state])
state = self.agent.GetNextState(state, action)
self.stateList.append(state)
self.maze[0][0] = 0.8
self.SetCanvas(self.maze)
self.timer.start(1000)
def DisplayNext(self):
if len(self.stateList) == 1:
QMessageBox.information(self, "Maze", "演示结束！", QMessageBox.Ok)
self.timer.stop()
return
state = self.stateList[1]
self.maze[state // self.agent.maze.m][state % self.agent.maze.m] = 0.8
self.maze[self.stateList[0] //
self.agent.maze.m][self.stateList[0] % self.agent.maze.m] = 1
self.SetCanvas(self.maze)
self.stateList.remove(self.stateList[0])
self.timer.start(1000)
def SetCanvas(self, maze):
plt.grid('on')
[nrows, ncols] = np.shape(maze)
ax = plt.gca()
ax.set_xticks(np.arange(0.5, nrows, 1))
ax.set_yticks(np.arange(0.5, ncols, 1))
ax.set_xticklabels([])
ax.set_yticklabels([])
canvas = np.copy(maze)
plt.imshow(canvas, interpolation='none', cmap='cubehelix')
self.canvas.draw()
epsilon -= 1 / (trainingTime // 2) # epsilon递减
# 随机初始化训练起点
state = np.random.randint(agentTraining.stateNum)
while mazeTraining.GetState(state) != 1:
state = np.random.randint(agentTraining.stateNum)
time = 0
while mazeTraining.GetState(state) != 0.9:
# epsilon贪心策略
if False not in (agentTraining.qTable[:, state] == [0, 0, 0, 0]):
action = np.random.randint(4)
else:
if random.random() < epsilon:
action = np.random.randint(4)
else:
action = np.argmax(agentTraining.qTable[:, state])
nextState = agentTraining.GetNextState(state, action)
# 更新Q表
agentTraining.qTable[action,state] =(1 - learningRate) * agentTraining.qTable[action,state] +\
learningRate * (agentTraining.maze.reward[action,state] + gamma * np.amax(agentTraining.qTable[:,nextState]))
state = nextState
time += 1
if time > 200: # 避免陷入同一个状态一致一直循环
break
if mazeTraining.GetState(state) == 0.5:
break
# 显示训练进度
if i % (trainingTime / 10) == 0 and i != 0:
print('episodes:' + str(i) + '/' + str(trainingTime))
print('episodes:' + str(trainingTime) + '/' + str(trainingTime))
print('训练完成！')
agentSaveFileName = input('请输入训练结果保存的文件名：')
