RL之Q Learning：利用强化学习之Q Learning实现走迷宫—训练智能体走到迷宫(复杂迷宫)的宝藏位置

输出结果

设计思路

实现代码

from __future__ import print_function

import numpy as np

import time

from env import Env

from reprint import output

EPSILON = 0.1

ALPHA = 0.1

GAMMA = 0.9

MAX_STEP = 30

np.random.seed(0)

def epsilon_greedy(Q, state):

   if (np.random.uniform() > 1 - EPSILON) or ((Q[state, :] == 0).all()):

       action = np.random.randint(0, 4)  # 0~3

   else:

       action = Q[state, :].argmax()

   return action

e = Env()

Q = np.zeros((e.state_num, 4))

with output(output_type="list", initial_len=len(e.map), interval=0) as output_list:

   for i in range(100):

       e = Env()

       while (e.is_end is False) and (e.step < MAX_STEP):

           action = epsilon_greedy(Q, e.present_state)

           state = e.present_state

           reward = e.interact(action)

           new_state = e.present_state

           Q[state, action] = (1 - ALPHA) * Q[state, action] + \

               ALPHA * (reward + GAMMA * Q[new_state, :].max())

           e.print_map_with_reprint(output_list)

           time.sleep(0.1)

       for line_num in range(len(e.map)):

           if line_num == 0:

               output_list[0] = 'Episode:{} Total Step:{}, Total Reward:{}'.format(i, e.step, e.total_reward)

           else:

               output_list[line_num] = ''

       time.sleep(2)