【零基础强化学习】 基于tkinter搭建gridworld强化学习环境

更多代码： Gitee主页：https://gitee.com/GZHzzz
博客主页： CSDN：https://blog.csdn.net/gzhzzaa

写在前面

• 作为一个新手，写这个强化学习-基础知识专栏是想和大家分享一下自己强化学习的学习历程，希望大家互相交流一起进步！😁在我的gitee收集了强化学习经典论文：强化学习经典论文，搭建了基于pytorch的典型智能体模型，大家一起多篇多交流，互相学习啊！😊
  

show me code, no bb

"""
Reinforcement learning maze example.
Red rectangle:          explorer.
Black rectangles:       hells       [reward = -1].
Yellow bin circle:      paradise    [reward = +1].
All other states:       ground      [reward = 0].
"""
import numpy as np
import pandas as pd
import numpy as np
import time
import sys
import tkinter as tk
 
UNIT = 40   # pixels   像素
MAZE_H = 10  # grid height
MAZE_W = 10  # grid width
 
class Maze(tk.Tk, object):
    def __init__(self):
        super(Maze, self).__init__()
        self.action_space = ['u', 'd', 'l', 'r']                #行为
        self.n_actions = len(self.action_space)                  #行为数
        self.title('maze')
        self.geometry('{0}x{1}'.format(MAZE_H * UNIT, MAZE_H * UNIT))
        self._build_maze()
 
    def _build_maze(self):
        self.canvas = tk.Canvas(self, bg='white',
                           height=MAZE_H * UNIT,
                           width=MAZE_W * UNIT)
 
        # create grids
        for c in range(0, MAZE_W * UNIT, UNIT):
            x0, y0, x1, y1 = c, 0, c, MAZE_W * UNIT
            self.canvas.create_line(x0, y0, x1, y1)   #画一条从(x0,y0)到(x1,y1)的线
        for r in range(0, MAZE_H * UNIT, UNIT):
            x0, y0, x1, y1 = 0, r, MAZE_H * UNIT, r
            self.canvas.create_line(x0, y0, x1, y1)
 
        # create origin
        origin = np.array([20, 20])
 
        # hell            #画第黑色正方形
        x = [
        1, 8, 5, 9, 2, 4, 3, 9, 8, 7, 5, 3, 1, 0, 4, 5,7, 9, 1,7,4,6,6,8]
        y = [
        2, 2, 9, 3, 0, 6, 0, 9, 0, 4, 4, 4, 9, 4 , 1, 2,4, 3, 6,7,8,0,6,5]
        self.res = []
        for i in range(24):
            oval_center= origin+np.array([x[i]*UNIT,y[i]*UNIT])
            self.rec = self.canvas.create_rectangle(
            oval_center[0] - 15, oval_center[1] - 15,
            oval_center[0] + 15, oval_center[1] + 15,
            fill='black')
            self.res.append(self.canvas.coords(self.rec))




        # create oval     #画黄色
        oval_center = origin + UNIT * 8
        self.oval = self.canvas.create_oval(
            oval_center[0] - 15, oval_center[1] - 15,
            oval_center[0] + 15, oval_center[1] + 15,
            fill='yellow')
 
        # create red rect   #画红色的正方形
        self.rect = self.canvas.create_rectangle(
            origin[0] - 15, origin[1] - 15,
            origin[0] + 15, origin[1] + 15,
            fill='red')
        # pack all
        self.canvas.pack()
 
    def reset(self):
        # self.update()
        # time.sleep(0.1)
        self.canvas.delete(self.rect)
        origin = np.array([20, 20])
        self.rect = self.canvas.create_rectangle(
            origin[0] - 15, origin[1] - 15,
            origin[0] + 15, origin[1] + 15,
            fill='red')
        # return observation
        return self.canvas.coords(self.rect)
 
    def step(self, action):
        s = self.canvas.coords(self.rect)
        base_action = np.array([0, 0])
        reward = 0
        done = False
        if action == 0:   # up
            if s[1] > UNIT:
                base_action[1] -= UNIT    #减40
            else:
                reward = -5
        elif action == 1:   # down
            if s[1] < (MAZE_H - 1) * UNIT:
                base_action[1] += UNIT     #加40
            else:
                reward = -5
        elif action == 2:   # right
            if s[0] < (MAZE_W - 1) * UNIT:
                base_action[0] += UNIT    #右移40
            else:
                reward = -5
        elif action == 3:   # left
            if s[0] > UNIT:               #左移40
                base_action[0] -= UNIT
            else:
                reward = -5
        self.canvas.move(self.rect, base_action[0], base_action[1])  # move agent
 
        s_ = self.canvas.coords(self.rect)  # next state
        # reward function
        if s_ == self.canvas.coords(self.oval):
            reward = 100
            done = True
            # s_ = 'terminal'
        if s_ in self.res:
            reward = -5
            
            self.canvas.move(self.rect, -base_action[0], -base_action[1])  # move agent
            s_ = self.canvas.coords(self.rect)  # next state
            
            # s_ = 'terminal'
        
        return s_, reward, done
 
    def render(self):
        time.sleep(0.1)
        # time.sleep(10)
        self.update()
 
if __name__ == "__main__":
    env = Maze()
    while True:
            # fresh env
        env.render()
    env.render()
  

    

• 代码全部亲自跑过，你懂的！😝

结果展示

• reward与状态转移可以由任务自适应改变

写在最后

十年磨剑，与君共勉！
更多代码：Gitee主页：https://gitee.com/GZHzzz
博客主页：CSDN：https://blog.csdn.net/gzhzzaa

• Fighting!😎

基于pytorch的经典模型：基于pytorch的典型智能体模型
强化学习经典论文：强化学习经典论文

while True:
	Go life

谢谢点赞交流！(❁´◡`❁)