def decode(self, obj: bytes) -> Message:
"""
Decode the message.
:param obj: the bytes object
:return: the message
"""
json_msg = json.loads(obj.decode("utf-8"))
performative = GymMessage.Performative(json_msg["performative"])
new_body = copy.copy(json_msg)
new_body["performative"] = performative
if performative == GymMessage.Performative.ACT:
action_bytes = base64.b64decode(json_msg["action"])
action = pickle.loads(action_bytes) # nosec
new_body["action"] = action
new_body["step_id"] = json_msg["step_id"]
elif performative == GymMessage.Performative.PERCEPT:
# observation, reward and info are gym implementation specific, done is boolean
observation_bytes = base64.b64decode(json_msg["observation"])
observation = pickle.loads(observation_bytes) # nosec
new_body["observation"] = observation
reward_bytes = base64.b64decode(json_msg["reward"])
reward = pickle.loads(reward_bytes) # nosec
new_body["reward"] = reward
info_bytes = base64.b64decode(json_msg["info"])
info = pickle.loads(info_bytes) # nosec
new_body["info"] = info
new_body["step_id"] = json_msg["step_id"]
gym_message = GymMessage(performative=performative, body=new_body)
return gym_message
def _decode_percept(self, envelope: Envelope,
expected_step_id: int) -> Message:
"""
Receive the response from the gym environment in the form of an envelope and decode it.
The response is a PERCEPT message containing the usual 'observation', 'reward', 'done', 'info' parameters.
:param expected_step_id: the expected step id
:return: a message received as a response to the action performed in apply_action.
"""
if envelope is not None:
if envelope.protocol_id == PublicId.from_str("fetchai/gym:0.1.0"):
gym_msg = GymSerializer().decode(envelope.message)
gym_msg_performative = GymMessage.Performative(
gym_msg.get("performative"))
gym_msg_step_id = gym_msg.get("step_id")
if (gym_msg_performative == GymMessage.Performative.PERCEPT
and gym_msg_step_id == expected_step_id):
return gym_msg
else:
raise ValueError(
"Unexpected performative or no step_id: {}".format(
gym_msg_performative))
else:
raise ValueError("Unknown protocol_id: {}".format(
envelope.protocol_id))
else:
raise ValueError("Missing envelope.")
def decode(obj: bytes) -> Message:
"""
Decode bytes into a 'Gym' message.
:param obj: the bytes object.
:return: the 'Gym' message.
"""
message_pb = ProtobufMessage()
gym_pb = gym_pb2.GymMessage()
message_pb.ParseFromString(obj)
message_id = message_pb.dialogue_message.message_id
dialogue_reference = (
message_pb.dialogue_message.dialogue_starter_reference,
message_pb.dialogue_message.dialogue_responder_reference,
)
target = message_pb.dialogue_message.target
gym_pb.ParseFromString(message_pb.dialogue_message.content)
performative = gym_pb.WhichOneof("performative")
performative_id = GymMessage.Performative(str(performative))
performative_content = dict() # type: Dict[str, Any]
if performative_id == GymMessage.Performative.ACT:
pb2_action = gym_pb.act.action
action = AnyObject.decode(pb2_action)
performative_content["action"] = action
step_id = gym_pb.act.step_id
performative_content["step_id"] = step_id
elif performative_id == GymMessage.Performative.PERCEPT:
step_id = gym_pb.percept.step_id
performative_content["step_id"] = step_id
pb2_observation = gym_pb.percept.observation
observation = AnyObject.decode(pb2_observation)
performative_content["observation"] = observation
reward = gym_pb.percept.reward
performative_content["reward"] = reward
done = gym_pb.percept.done
performative_content["done"] = done
pb2_info = gym_pb.percept.info
info = AnyObject.decode(pb2_info)
performative_content["info"] = info
elif performative_id == GymMessage.Performative.STATUS:
content = gym_pb.status.content
content_dict = dict(content)
performative_content["content"] = content_dict
elif performative_id == GymMessage.Performative.RESET:
pass
elif performative_id == GymMessage.Performative.CLOSE:
pass
else:
raise ValueError(
"Performative not valid: {}.".format(performative_id))
return GymMessage(message_id=message_id,
dialogue_reference=dialogue_reference,
target=target,
performative=performative,
**performative_content)
def handle_gym_message(self, envelope: Envelope) -> None:
"""
Forward a message to gym.
:param envelope: the envelope
:return: None
"""
gym_message = GymSerializer().decode(envelope.message)
performative = gym_message.get("performative")
if GymMessage.Performative(
performative) == GymMessage.Performative.ACT:
action = gym_message.get("action")
step_id = gym_message.get("step_id")
observation, reward, done, info = self.gym_env.step(
action) # type: ignore
msg = GymMessage(
performative=GymMessage.Performative.PERCEPT,
observation=observation,
reward=reward,
done=done,
info=info,
step_id=step_id,
)
msg_bytes = GymSerializer().encode(msg)
envelope = Envelope(
to=envelope.sender,
sender=DEFAULT_GYM,
protocol_id=GymMessage.protocol_id,
message=msg_bytes,
)
self._send(envelope)
elif GymMessage.Performative(
performative) == GymMessage.Performative.RESET:
self.gym_env.reset() # type: ignore
elif GymMessage.Performative(
performative) == GymMessage.Performative.CLOSE:
self.gym_env.close() # type: ignore