One bug in your codes

[ZiyiLiubird]

Hello, thanks for your contribution to this excellent work. I find perhaps one bug in your code, and the link is as follows:

QDAgents/qd_agents/networks/dice_manager.py

Line 102 in 120214f

self.dratios[i].predict_batch_rewards(i, self.rl_agent.rollouts, self.inverse_dratios[i], self.rl_agent.get_action)

Maybe this line of code should be in one loop, e.g., for i in self.neighbors.
I'm looking forward to your replay, thanks!