强化学习Q-Learning解决FrozenLake例子(Python)
import gym
import numpy as np
import random
import matplotlib.pyplot as plt
# gym创建冰湖环境
env = gym.make('FrozenLake-v0')
# 初始化Q表格,矩阵维度为【S,A】,即状态数*动作数
Q_all = np.zeros([env.observation_space.n,env....
原创
2019-01-29 17:32:15 ·
7441 阅读 ·
4 评论