def step(self, action):
utils.send_message(self.sock,
gym_uds_pb2.Request(type=gym_uds_pb2.Request.STEP))
utils.send_message(self.sock, gym_uds_pb2.Action(value=action))
state_pb = utils.recv_message(self.sock, gym_uds_pb2.State)
observation = np.asarray(state_pb.observation.data).reshape(
state_pb.observation.shape)
return observation, state_pb.reward, state_pb.done
def Sample(self, empty_request, context):
action = self.env.action_space.sample()
if (isinstance(action, (np.ndarray, np.generic))):
act = action.tolist()
elif (isinstance(action, tuple)):
act = [x for x in action]
elif (isinstance(action, int)):
act = [float(action)]
elif (isinstance(action, float)):
act = [action]
return gym_uds_pb2.Action(data=act)
def Sample(self, empty_request, context):
action = self.env.action_space.sample()
return gym_uds_pb2.Action(value=action)
def step(self, action):
state_pb = self.stub.Step(gym_uds_pb2.Action(value=action))
observation = np.asarray(state_pb.observation.data).reshape(
state_pb.observation.shape)
return observation, state_pb.reward, state_pb.done
def sample(self):
action = self.env.action_space.sample()
utils.send_message(self.sock, gym_uds_pb2.Action(value=action))