ModelBasedEnvBase

torchrl.envs.ModelBasedEnvBase(*args, **kwargs)

基于模型的 RL sota 实现的基本环境。

围绕 MBRL 算法模型的包装器。它旨在为世界模型提供环境框架（包括但不限于观测、奖励、完成状态和安全约束模型），并表现得像一个经典环境。

这是其他环境的基类，不应直接使用。

示例

>>> import torch
>>> from tensordict import TensorDict
>>> from torchrl.data import Composite, Unbounded
>>> class MyMBEnv(ModelBasedEnvBase):
...     def __init__(self, world_model, device="cpu", dtype=None, batch_size=None):
...         super().__init__(world_model, device=device, dtype=dtype, batch_size=batch_size)
...         self.observation_spec = Composite(
...             hidden_observation=Unbounded((4,))
...         )
...         self.state_spec = Composite(
...             hidden_observation=Unbounded((4,)),
...         )
...         self.action_spec = Unbounded((1,))
...         self.reward_spec = Unbounded((1,))
...
...     def _reset(self, tensordict: TensorDict) -> TensorDict:
...         tensordict = TensorDict({},
...             batch_size=self.batch_size,
...             device=self.device,
...         )
...         tensordict = tensordict.update(self.state_spec.rand())
...         tensordict = tensordict.update(self.observation_spec.rand())
...         return tensordict
>>> # This environment is used as follows:
>>> import torch.nn as nn
>>> from torchrl.modules import MLP, WorldModelWrapper
>>> world_model = WorldModelWrapper(
...     TensorDictModule(
...         MLP(out_features=4, activation_class=nn.ReLU, activate_last_layer=True, depth=0),
...         in_keys=["hidden_observation", "action"],
...         out_keys=["hidden_observation"],
...     ),
...     TensorDictModule(
...         nn.Linear(4, 1),
...         in_keys=["hidden_observation"],
...         out_keys=["reward"],
...     ),
... )
>>> env = MyMBEnv(world_model)
>>> tensordict = env.rollout(max_steps=10)
>>> print(tensordict)
TensorDict(
    fields={
        action: Tensor(torch.Size([10, 1]), dtype=torch.float32),
        done: Tensor(torch.Size([10, 1]), dtype=torch.bool),
        hidden_observation: Tensor(torch.Size([10, 4]), dtype=torch.float32),
        next: LazyStackedTensorDict(
            fields={
                hidden_observation: Tensor(torch.Size([10, 4]), dtype=torch.float32)},
            batch_size=torch.Size([10]),
            device=cpu,
            is_shared=False),
        reward: Tensor(torch.Size([10, 1]), dtype=torch.float32)},
    batch_size=torch.Size([10]),
    device=cpu,
    is_shared=False)

属性

• observation_spec (Composite): 观测的采样规范；

• action_spec (TensorSpec): 动作的采样规范；

• reward_spec (TensorSpec): 奖励的采样规范；

• input_spec (Composite): 输入的采样规范；

• batch_size (torch.Size): env 要使用的 batch_size。如果未设置，则 env 接受所有 batch_size 的 tensordict。

• device (torch.device): env 输入和输出预期存在的设备

参数:

• world_model (nn.Module) – 生成世界状态及其相应奖励的模型；

• params (List[torch.Tensor], 可选) – 世界模型的参数列表；

• buffers (List[torch.Tensor], 可选) – 世界模型的缓冲区列表；

• device (torch.device, 可选) – env 输入和输出预期存在的设备

• dtype (torch.dtype, 可选) – env 输入和输出的 dtype

• batch_size (torch.Size, 可选) – 实例中包含的环境数量

• run_type_check (bool, 可选) – 是否在 env 的步骤中运行类型检查

torchrl.envs.step(TensorDict -> TensorDict)

环境中的步骤

torchrl.envs.reset(TensorDict, optional -> TensorDict)

重置环境

torchrl.envs.set_seed(int -> int)

设置环境的种子

torchrl.envs.rand_step(TensorDict, optional -> TensorDict)

给定动作规范的随机步骤

torchrl.envs.rollout(Callable, ... -> TensorDict)

使用给定的策略在环境中执行 rollout（如果未提供策略，则执行随机步骤）