#observation_dim = 9
# 定义两组不同的事件，update 和 rolling
UPDATE_EVENT = Event() # ppo更新事件
ROLLING_EVENT = Event() # worker收集数据事件
UPDATE_EVENT.clear() # not update now，相当于把标志位设置为False ppo事件停止
ROLLING_EVENT.set(
) # start to roll out，相当于把标志位设置为True，并通知所有处于等待阻塞状态的线程恢复运行状态。 worker开始工作
update_rlock = RLock()
episode_rlock = RLock()
ns = Manager().Namespace()
ns.GLOBAL_UPDATE_COUNTER = 0
ns.GLOBAL_EPISODE = 0
ns.GLOBAL_RUNNING_REWARD = []
ns.coord_status = True
QUEUE = Manager().Queue()
class MyManager(BaseManager):
pass
MyManager.register('Ppo', PPO)
manager = MyManager()
manager.start()
Ppo = manager.Ppo(action_dim, observation_dim, is_train=True)
# GLOBAL_PPO = PPO(action_dim, observation_dim, is_train=True) #一个global的ppo
print('<TRAIN_LOG> ', 'GLOBAL_PPO get', ' time:', datetime.now())
try:
