diff --git a/reinforcement_learning/ppo/ppo_agent.py b/reinforcement_learning/ppo/ppo_agent.py index 21ce12d3653e7042e6e7f7585ff1cfaa0d2da104..54ba25fe1da709ca26587193561ac2e0a3b8b4a9 100644 --- a/reinforcement_learning/ppo/ppo_agent.py +++ b/reinforcement_learning/ppo/ppo_agent.py @@ -35,7 +35,8 @@ class DataBuffers: return self.memory.get(handle, []) def push_transition(self, handle, transition): - self.memory.update({handle: self.get_transitions(handle).append(transition)}) + transitions = self.get_transitions(handle).append(transition) + self.memory.update({handle: transitions}) class PPOAgent(Policy):