def learning(best_agent):
# print_experience
best_agent.print_experience()
print_state(best_agent.get_q_values)
agent = QLearning(12, 4)
# 学習前
agent.episodes = best_agent.episodes
# 学習
for i, episode in enumerate(best_agent.episodes):
agent.history += episode
if i % 10 == 0 and not i == 0:
agent.learn()
agent.reset()
# 学習前
print_state(agent.get_q_values)
# print_experience
agent.print_experience()
def learning(best_agent):
# print_experience
best_agent.print_experience()
print_state(best_agent.get_q_values)
agent = QLearning(12, 4)
# 学習前
agent.episodes = best_agent.episodes
# 学習
for i, episode in enumerate(best_agent.episodes):
agent.history += episode
if i % 10 == 0 and not i == 0:
agent.learn()
agent.reset()
# 学習前
print_state(agent.get_q_values)
# print_experience
agent.print_experience()
