verl/workers/actor/base.py

• BasePPOActor · class · L26-L66 — class BasePPOActor(ABC)
• init · method · L28-L36 — def init(self, config)
• compute_log_prob · method · L39-L51 — def compute_log_prob(self, data: DataProto) -> torch.Tensor
• update_policy · method · L54-L66 — def update_policy(self, data: DataProto) -> Dict