Closed hmzo closed 5 months ago
The reason seems to be that the reference occurs before the assignment in these two lines:
https://github.com/OpenLLMAI/OpenRLHF/blob/main/openrlhf/trainer/ray/ppo_actor.py#L194 https://github.com/OpenLLMAI/OpenRLHF/blob/main/openrlhf/trainer/ray/ppo_critic.py#L98
Fixed https://github.com/OpenLLMAI/OpenRLHF/pull/197
The reason seems to be that the reference occurs before the assignment in these two lines:
https://github.com/OpenLLMAI/OpenRLHF/blob/main/openrlhf/trainer/ray/ppo_actor.py#L194 https://github.com/OpenLLMAI/OpenRLHF/blob/main/openrlhf/trainer/ray/ppo_critic.py#L98