if action is not None:
reward = get_reward(status)
reward_printer(state, action, reward)
q_learner.update(state, action, reward)
action = q_learner.get_action(state, valid_teammates)
if action == 0:
print("Action Taken: DRIBBLE \n")
hfo.act(DRIBBLE)
elif action == 1:
print("Action Taken: SHOOT \n")
hfo.act(SHOOT)
elif args.numTeammates > 0:
print("Action Taken: PASS -> {0} \n".format(action - 2))
hfo.act(PASS, features[15 + 6 * (action - 2)])
status = hfo.step()
if action is not None and state is not None:
reward = get_reward(status)
reward_printer(state, action, reward)
q_learner.update(state, action, reward)
q_learner.clear()
q_learner.save()
if status == SERVER_DOWN:
hfo.act(QUIT)
q_learner.save()
break
q_learner.save()
