RobosetExperienceReplay¶

類 torchrl.data.datasets.RobosetExperienceReplay(dataset_id, batch_size: int, *, root: str | Path | None = None, download: bool = True, sampler: Sampler | None = None, writer: Writer | None = None, collate_fn: Callable | None = None, pin_memory: bool = False, prefetch: int | None = None, transform: 'torchrl.envs.Transform' | None = None, split_trajs: bool = False, **env_kwargs)[source]¶

Roboset 經驗回放資料集。

此類從 Roboset 下載 H5 資料，並以 mmap 格式進行處理，從而加快了索引（以及取樣）速度。

在此瞭解更多關於 Roboset 的資訊：https://sites.google.com/view/robohive/roboset

資料格式遵循 TED 約定。

引數:

dataset_id (str) – 要下載的資料集。必須是 RobosetExperienceReplay.available_datasets 的一部分。
batch_size (int) – 取樣期間使用的批次大小。如有必要，可以透過 data.sample(batch_size) 覆蓋此值。

關鍵字引數:

root (Path 或 str, 可選) – Roboset 資料集的根目錄。實際的資料集記憶體對映檔案將儲存在 <root>/<dataset_id> 下。如果未提供，預設路徑為 ~/.cache/torchrl/atari.roboset`。
download (bool 或 str, 可選) – 如果未找到資料集，是否應該下載。預設為 True。下載選項也可以傳遞 "force"，在這種情況下，已下載的資料將被覆蓋。
sampler (Sampler, 可選) – 要使用的取樣器。如果未提供，將使用預設的 RandomSampler()。
writer (Writer, 可選) – 要使用的寫入器。如果未提供，將使用預設的 ImmutableDatasetWriter。
collate_fn (callable, 可選) – 合併樣本列表以形成 Tensor(s)/輸出的 mini-batch。在從 map-style 資料集進行批次載入時使用。
pin_memory (bool) – 是否應對回放緩衝區樣本呼叫 pin_memory()。
prefetch (int, 可選) – 使用多執行緒預取後續批次的數量。
transform (Transform, 可選) – 在呼叫 sample() 時執行的轉換。要鏈式使用轉換，請使用 Compose 類。
split_trajs (bool, 可選) – 如果為 True，軌跡將沿第一維度拆分並填充以具有匹配的形狀。為了拆分軌跡，將使用 "done" 訊號，該訊號透過 done = truncated | terminated 恢復。換句話說，假定任何 truncated 或 terminated 訊號都等同於軌跡的結束。預設為 False。

變數:

available_datasets – 可接受的、可供下載的條目列表。

示例

>>> import torch
>>> torch.manual_seed(0)
>>> from torchrl.envs.transforms import ExcludeTransform
>>> from torchrl.data.datasets import RobosetExperienceReplay
>>> d = RobosetExperienceReplay("FK1-v4(expert)/FK1_MicroOpenRandom_v2d-v4", batch_size=32,
...     transform=ExcludeTransform("info", ("next", "info")))  # excluding info dict for conciseness
>>> for batch in d:
...     break
>>> # data is organised by seed and episode, but stored contiguously
>>> print(f"{batch['seed']}, {batch['episode']}")
tensor([2, 1, 0, 0, 1, 1, 0, 0, 1, 1, 2, 2, 2, 2, 2, 1, 1, 2, 0, 2, 0, 2, 2, 1,
        0, 2, 0, 0, 1, 1, 2, 1]) tensor([17, 20, 18,  9,  6,  1, 12,  6,  2,  6,  8, 15,  8, 21, 17,  3,  9, 20,
        23, 12,  3, 16, 19, 16, 16,  4,  4, 12,  1,  2, 15, 24])
>>> print(batch)
TensorDict(
    fields={
        action: Tensor(shape=torch.Size([32, 9]), device=cpu, dtype=torch.float64, is_shared=False),
        done: Tensor(shape=torch.Size([32, 1]), device=cpu, dtype=torch.bool, is_shared=False),
        episode: Tensor(shape=torch.Size([32]), device=cpu, dtype=torch.int64, is_shared=False),
        index: Tensor(shape=torch.Size([32]), device=cpu, dtype=torch.int64, is_shared=False),
        next: TensorDict(
            fields={
                done: Tensor(shape=torch.Size([32, 1]), device=cpu, dtype=torch.bool, is_shared=False),
                observation: Tensor(shape=torch.Size([32, 75]), device=cpu, dtype=torch.float64, is_shared=False),
                reward: Tensor(shape=torch.Size([32, 1]), device=cpu, dtype=torch.float64, is_shared=False),
                terminated: Tensor(shape=torch.Size([32, 1]), device=cpu, dtype=torch.bool, is_shared=False),
                truncated: Tensor(shape=torch.Size([32, 1]), device=cpu, dtype=torch.bool, is_shared=False)},
            batch_size=torch.Size([32]),
            device=cpu,
            is_shared=False),
        observation: Tensor(shape=torch.Size([32, 75]), device=cpu, dtype=torch.float64, is_shared=False),
        seed: Tensor(shape=torch.Size([32]), device=cpu, dtype=torch.int64, is_shared=False),
        terminated: Tensor(shape=torch.Size([32, 1]), device=cpu, dtype=torch.bool, is_shared=False),
        time: Tensor(shape=torch.Size([32]), device=cpu, dtype=torch.float64, is_shared=False)},
        truncated: Tensor(shape=torch.Size([32, 1]), device=cpu, dtype=torch.bool, is_shared=False)},
    batch_size=torch.Size([32]),
    device=cpu,
    is_shared=False)

add(data: TensorDictBase) → int¶

向回放緩衝區新增單個元素。

引數:: data (Any) – 要新增到回放緩衝區的資料
返回:: 資料在回放緩衝區中的索引。

append_transform(transform: Transform, *, invert: bool = False) → ReplayBuffer¶

將轉換新增到末尾。

呼叫 sample 時，轉換按順序應用。

引數:: transform (Transform) – 要新增的轉換
關鍵字引數:: invert (bool, 可選) – 如果為 True，則轉換將被反轉（寫入期間呼叫正向，讀取期間呼叫逆向）。預設為 False。

示例

>>> rb = ReplayBuffer(storage=LazyMemmapStorage(10), batch_size=4)
>>> data = TensorDict({"a": torch.zeros(10)}, [10])
>>> def t(data):
...     data += 1
...     return data
>>> rb.append_transform(t, invert=True)
>>> rb.extend(data)
>>> assert (data == 1).all()

屬性 data_path¶: 資料集的路徑，包括分片。

屬性 data_path_root¶: 資料集根目錄的路徑。

delete()¶: 從磁碟刪除資料集儲存。

dump(*args, **kwargs)¶: dumps() 的別名。

dumps(path)¶

將回放緩衝區儲存到磁碟上的指定路徑。

引數:: path (Path 或 str) – 儲存回放緩衝區的路徑。

示例

>>> import tempfile
>>> import tqdm
>>> from torchrl.data import LazyMemmapStorage, TensorDictReplayBuffer
>>> from torchrl.data.replay_buffers.samplers import PrioritizedSampler, RandomSampler
>>> import torch
>>> from tensordict import TensorDict
>>> # Build and populate the replay buffer
>>> S = 1_000_000
>>> sampler = PrioritizedSampler(S, 1.1, 1.0)
>>> # sampler = RandomSampler()
>>> storage = LazyMemmapStorage(S)
>>> rb = TensorDictReplayBuffer(storage=storage, sampler=sampler)
>>>
>>> for _ in tqdm.tqdm(range(100)):
...     td = TensorDict({"obs": torch.randn(100, 3, 4), "next": {"obs": torch.randn(100, 3, 4)}, "td_error": torch.rand(100)}, [100])
...     rb.extend(td)
...     sample = rb.sample(32)
...     rb.update_tensordict_priority(sample)
>>> # save and load the buffer
>>> with tempfile.TemporaryDirectory() as tmpdir:
...     rb.dumps(tmpdir)
...
...     sampler = PrioritizedSampler(S, 1.1, 1.0)
...     # sampler = RandomSampler()
...     storage = LazyMemmapStorage(S)
...     rb_load = TensorDictReplayBuffer(storage=storage, sampler=sampler)
...     rb_load.loads(tmpdir)
...     assert len(rb) == len(rb_load)

empty()¶: 清空回放緩衝區並將遊標重置為 0。

extend(tensordicts: TensorDictBase) → Tensor¶

使用可迭代物件中包含的一個或多個元素擴充套件回放緩衝區。

如果存在逆向轉換，將呼叫它們。

引數:: data (iterable) – 要新增到回放緩衝區的資料集合。
返回:: 新增到回放緩衝區的資料的索引。

警告

extend() 在處理值列表時可能存在模糊的簽名，這些值列表應被解釋為 PyTree（在這種情況下，列表中的所有元素都將放入儲存中 PyTree 的一個切片中），或者被解釋為要逐個新增的值列表。為了解決這個問題，TorchRL 對 list 和 tuple 進行了明確區分：tuple 將被視為 PyTree，而 list（在根級別）將被解釋為要逐個新增到緩衝區的值堆疊。對於 ListStorage 例項，只能提供未繫結元素（非 PyTree）。

insert_transform(index: int, transform: Transform, *, invert: bool = False) → ReplayBuffer¶

插入轉換。

呼叫 sample 時，轉換按順序執行。

引數:

index (int) – 插入轉換的位置。
transform (Transform) – 要新增的轉換

關鍵字引數:

invert (bool, 可選) – 如果為 True，則轉換將被反轉（寫入期間呼叫正向，讀取期間呼叫逆向）。預設為 False。

load(*args, **kwargs)¶: loads() 的別名。

loads(path)¶

在給定路徑載入回放緩衝區狀態。

緩衝區應具有匹配的元件，並使用 dumps() 儲存。

引數:: path (Path 或 str) – 儲存回放緩衝區的路徑。

有關更多資訊，請參閱 dumps()。

preprocess(fn: Callable[[TensorDictBase], TensorDictBase], dim: int = 0, num_workers: int | None = None, *, chunksize: int | None = None, num_chunks: int | None = None, pool: mp.Pool | None = None, generator: torch.Generator | None = None, max_tasks_per_child: int | None = None, worker_threads: int = 1, index_with_generator: bool = False, pbar: bool = False, mp_start_method: str | None = None, num_frames: int | None = None, dest: str | Path) → TensorStorage¶

預處理資料集並返回帶有格式化資料的新儲存。

資料轉換必須是單一的（作用於資料集中的單個樣本）。

引數和關鍵字引數將轉發到 map()。

隨後可以使用 delete() 刪除資料集。

關鍵字引數:

dest (路徑或 等效型別) – 新資料集位置的路徑。
num_frames (int, 可選) – 如果提供，僅轉換前 num_frames 幀。這對於初步除錯轉換很有用。

返回: 一個新的儲存，用於 ReplayBuffer 例項內部。

示例

>>> from torchrl.data.datasets import MinariExperienceReplay
>>>
>>> data = MinariExperienceReplay(
...     list(MinariExperienceReplay.available_datasets)[0],
...     batch_size=32
...     )
>>> print(data)
MinariExperienceReplay(
    storages=TensorStorage(TensorDict(
        fields={
            action: MemoryMappedTensor(shape=torch.Size([1000000, 8]), device=cpu, dtype=torch.float32, is_shared=True),
            episode: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.int64, is_shared=True),
            info: TensorDict(
                fields={
                    distance_from_origin: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                    forward_reward: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                    goal: MemoryMappedTensor(shape=torch.Size([1000000, 2]), device=cpu, dtype=torch.float64, is_shared=True),
                    qpos: MemoryMappedTensor(shape=torch.Size([1000000, 15]), device=cpu, dtype=torch.float64, is_shared=True),
                    qvel: MemoryMappedTensor(shape=torch.Size([1000000, 14]), device=cpu, dtype=torch.float64, is_shared=True),
                    reward_ctrl: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                    reward_forward: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                    reward_survive: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                    success: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.bool, is_shared=True),
                    x_position: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                    x_velocity: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                    y_position: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                    y_velocity: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True)},
                batch_size=torch.Size([1000000]),
                device=cpu,
                is_shared=False),
            next: TensorDict(
                fields={
                    done: MemoryMappedTensor(shape=torch.Size([1000000, 1]), device=cpu, dtype=torch.bool, is_shared=True),
                    info: TensorDict(
                        fields={
                            distance_from_origin: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                            forward_reward: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                            goal: MemoryMappedTensor(shape=torch.Size([1000000, 2]), device=cpu, dtype=torch.float64, is_shared=True),
                            qpos: MemoryMappedTensor(shape=torch.Size([1000000, 15]), device=cpu, dtype=torch.float64, is_shared=True),
                            qvel: MemoryMappedTensor(shape=torch.Size([1000000, 14]), device=cpu, dtype=torch.float64, is_shared=True),
                            reward_ctrl: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                            reward_forward: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                            reward_survive: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                            success: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.bool, is_shared=True),
                            x_position: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                            x_velocity: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                            y_position: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True),
                            y_velocity: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.float64, is_shared=True)},
                        batch_size=torch.Size([1000000]),
                        device=cpu,
                        is_shared=False),
                    observation: TensorDict(
                        fields={
                            achieved_goal: MemoryMappedTensor(shape=torch.Size([1000000, 2]), device=cpu, dtype=torch.float64, is_shared=True),
                            desired_goal: MemoryMappedTensor(shape=torch.Size([1000000, 2]), device=cpu, dtype=torch.float64, is_shared=True),
                            observation: MemoryMappedTensor(shape=torch.Size([1000000, 27]), device=cpu, dtype=torch.float64, is_shared=True)},
                        batch_size=torch.Size([1000000]),
                        device=cpu,
                        is_shared=False),
                    reward: MemoryMappedTensor(shape=torch.Size([1000000, 1]), device=cpu, dtype=torch.float64, is_shared=True),
                    terminated: MemoryMappedTensor(shape=torch.Size([1000000, 1]), device=cpu, dtype=torch.bool, is_shared=True),
                    truncated: MemoryMappedTensor(shape=torch.Size([1000000, 1]), device=cpu, dtype=torch.bool, is_shared=True)},
                batch_size=torch.Size([1000000]),
                device=cpu,
                is_shared=False),
            observation: TensorDict(
                fields={
                    achieved_goal: MemoryMappedTensor(shape=torch.Size([1000000, 2]), device=cpu, dtype=torch.float64, is_shared=True),
                    desired_goal: MemoryMappedTensor(shape=torch.Size([1000000, 2]), device=cpu, dtype=torch.float64, is_shared=True),
                    observation: MemoryMappedTensor(shape=torch.Size([1000000, 27]), device=cpu, dtype=torch.float64, is_shared=True)},
                batch_size=torch.Size([1000000]),
                device=cpu,
                is_shared=False)},
        batch_size=torch.Size([1000000]),
        device=cpu,
        is_shared=False)),
    samplers=RandomSampler,
    writers=ImmutableDatasetWriter(),
batch_size=32,
transform=Compose(
),
collate_fn=<function _collate_id at 0x120e21dc0>)
>>> from torchrl.envs import CatTensors, Compose
>>> from tempfile import TemporaryDirectory
>>>
>>> cat_tensors = CatTensors(
...     in_keys=[("observation", "observation"), ("observation", "achieved_goal"),
...              ("observation", "desired_goal")],
...     out_key="obs"
...     )
>>> cat_next_tensors = CatTensors(
...     in_keys=[("next", "observation", "observation"),
...              ("next", "observation", "achieved_goal"),
...              ("next", "observation", "desired_goal")],
...     out_key=("next", "obs")
...     )
>>> t = Compose(cat_tensors, cat_next_tensors)
>>>
>>> def func(td):
...     td = td.select(
...         "action",
...         "episode",
...         ("next", "done"),
...         ("next", "observation"),
...         ("next", "reward"),
...         ("next", "terminated"),
...         ("next", "truncated"),
...         "observation"
...         )
...     td = t(td)
...     return td
>>> with TemporaryDirectory() as tmpdir:
...     new_storage = data.preprocess(func, num_workers=4, pbar=True, mp_start_method="fork", dest=tmpdir)
...     rb = ReplayBuffer(storage=new_storage)
...     print(rb)
ReplayBuffer(
    storage=TensorStorage(
        data=TensorDict(
            fields={
                action: MemoryMappedTensor(shape=torch.Size([1000000, 8]), device=cpu, dtype=torch.float32, is_shared=True),
                episode: MemoryMappedTensor(shape=torch.Size([1000000]), device=cpu, dtype=torch.int64, is_shared=True),
                next: TensorDict(
                    fields={
                        done: MemoryMappedTensor(shape=torch.Size([1000000, 1]), device=cpu, dtype=torch.bool, is_shared=True),
                        obs: MemoryMappedTensor(shape=torch.Size([1000000, 31]), device=cpu, dtype=torch.float64, is_shared=True),
                        observation: TensorDict(
                            fields={
                            },
                            batch_size=torch.Size([1000000]),
                            device=cpu,
                            is_shared=False),
                        reward: MemoryMappedTensor(shape=torch.Size([1000000, 1]), device=cpu, dtype=torch.float64, is_shared=True),
                        terminated: MemoryMappedTensor(shape=torch.Size([1000000, 1]), device=cpu, dtype=torch.bool, is_shared=True),
                        truncated: MemoryMappedTensor(shape=torch.Size([1000000, 1]), device=cpu, dtype=torch.bool, is_shared=True)},
                    batch_size=torch.Size([1000000]),
                    device=cpu,
                    is_shared=False),
                obs: MemoryMappedTensor(shape=torch.Size([1000000, 31]), device=cpu, dtype=torch.float64, is_shared=True),
                observation: TensorDict(
                    fields={
                    },
                    batch_size=torch.Size([1000000]),
                    device=cpu,
                    is_shared=False)},
            batch_size=torch.Size([1000000]),
            device=cpu,
            is_shared=False),
        shape=torch.Size([1000000]),
        len=1000000,
        max_size=1000000),
    sampler=RandomSampler(),
    writer=RoundRobinWriter(cursor=0, full_storage=True),
    batch_size=None,
    collate_fn=<function _collate_id at 0x168406fc0>)

register_load_hook(hook: Callable[[Any], Any])¶: 為儲存註冊載入鉤子。

注意

鉤子目前在儲存回放緩衝區時不會被序列化：它們必須在每次建立緩衝區時手動重新初始化。

register_save_hook(hook: Callable[[Any], Any])¶: 為儲存註冊儲存鉤子。

注意

鉤子目前在儲存回放緩衝區時不會被序列化：它們必須在每次建立緩衝區時手動重新初始化。

sample(batch_size: int | None = None, return_info: bool = False, include_info: bool = None) → TensorDictBase¶

從回放緩衝區取樣一批資料。

使用 Sampler 取樣索引，並從 Storage 中檢索它們。

引數:

batch_size (int, 可選) – 要收集的資料大小。如果未提供，此方法將按照取樣器指示的批次大小進行取樣。
return_info (bool) – 是否返回資訊。如果為 True，結果是一個元組 (data, info)。如果為 False，結果是資料。

返回:

包含在回放緩衝區中選擇的一批資料的 tensordict。如果 return_info 標誌設定為 True，則返回包含此 tensordict 和資訊的元組。

屬性 sampler¶

回放緩衝區的取樣器。

取樣器必須是 Sampler 的例項。

save(*args, **kwargs)¶: dumps() 的別名。

set_sampler(sampler: Sampler)¶: 在回放緩衝區中設定新的取樣器並返回之前的取樣器。

set_storage(storage: Storage, collate_fn: Callable | None = None)¶

在回放緩衝區中設定新的儲存器並返回之前的儲存器。

引數:

storage (Storage) – 緩衝區的新儲存器。
collate_fn (可呼叫物件, 可選) – 如果提供，collate_fn 將設定為此值。否則，它會重置為預設值。

set_writer(writer: Writer)¶: 在回放緩衝區中設定新的寫入器並返回之前的寫入器。

property storage¶

回放緩衝區的儲存器。

儲存器必須是 Storage 的例項。

property write_count¶: 到目前為止透過 add 和 extend 方法寫入緩衝區中的專案總數。

property writer¶

回放緩衝區的寫入器。

寫入器必須是 Writer 的例項。

RobosetExperienceReplay¶

文件

教程

資源