q_mask[2] -= 1000
actions_value_mask = [0, 0, 0, 0, 0]
for i in range(len(actions_value)):
actions_value_mask[i] = actions_value[i] + q_mask[i]
return actions_value_mask
logger = logging.getLogger(__name__)
logger.setLevel(level=logging.INFO)
handler = logging.FileHandler("log.txt")
handler.setLevel(logging.INFO)
logger.addHandler(handler)
my_env = env.TrafficEnv()
#bt.saver.restore(bt.sess,"./model/my_light_model/my-model.ckpt-3500")
reward_list = []
for i_episode in range(1000000):
# listener()
s = my_env.reset()
N_others = 12 * 10
s_pre_others = np.zeros((N_others))
s_pre_others2 = np.array(s[1] + s[2])
#print(s_pre_others2)
for i in range(N_others):
s_pre_others[i] = s_pre_others2[i % 12]
s_sliding, s_others = s[0], s_pre_others
#print(s_sliding,s_others)
import env
import time
import random
myenv = env.TrafficEnv()
myenv.reset()
while(1):
action = random.randint(0,4)
s,r,end,dis = myenv.step(action)
#time.sleep(1)
if end == 1:
myenv.reset()
print("end! ")
elif end == 100:
myenv.reset()
print("arrive!")
