RandomPolicy

torchrl.envs.utils.RandomPolicy(action_spec: TensorSpec, action_key: NestedKey = 'action')

用於資料收集器的隨機策略。

這是對 action_spec.rand 方法的包裝。

引數:

action_spec – 描述動作規範的 TensorSpec 物件

示例

>>> from tensordict import TensorDict
>>> from torchrl.data.tensor_specs import Bounded
>>> action_spec = Bounded(-torch.ones(3), torch.ones(3))
>>> actor = RandomPolicy(action_spec=action_spec)
>>> td = actor(TensorDict()) # selects a random action in the cube [-1; 1]