Language Language English 简体中文

unilab.algos.torch.offpolicy.worker.resolve_collector_actor_dims

unilab.algos.torch.offpolicy.worker.resolve_collector_actor_dims(env, obs_dim=None, action_dim=None)

Resolve actor dims for the collector.

Prefer explicit dims from the parent process so learner and collector build identical actor shapes on override-heavy env paths.

Parameters:

• obs_dim (int | None)

• action_dim (int | None)

Return type:

tuple[int, int]