GenDGRLExperienceReplay¶
- class torchrl.data.datasets.GenDGRLExperienceReplay(dataset_id: str, batch_size: int | None = None, *, download: bool = True, root: str | None = None, **kwargs)[原始碼]¶
Gen-DGRL 經驗回放資料集。
該資料集伴隨論文“The Generalization Gap in Offline Reinforcement Learning”。
Arxiv: https://arxiv.org/abs/2312.05742
GitHub: https://github.com/facebookresearch/gen_dgrl
資料格式遵循 TED 約定。
此類可讓您訪問 ProcGen 資料集。在 GenDGRLExperienceReplay.available_datasets 中註冊的每個 dataset_id 都由一個特定的任務(“bigfish”、“bossfight” 等)和一個類別(“1M_E”、“1M_S” 等)組成,它們用逗號分隔(“bigfish-1M_E”、...)。
在下載和準備過程中,資料會作為 .tar 檔案下載,其中每個軌跡都獨立儲存在 .npy 檔案中。這些檔案中的每一個都會被解壓,寫入一個連續的 mmap 張量,然後被清除。此過程可能需要幾分鐘才能處理完一個數據集。在叢集上,建議首先在不同的工作程序或程序上分別下載和預處理不同的資料集,然後在第二次啟動訓練指令碼。
- 引數:
dataset_id (str) – 要下載的資料集。必須是
GenDGRLExperienceReplay.available_datasets的一部分。batch_size (int, optional) – 取樣過程中使用的批次大小。如果需要,可以透過 data.sample(batch_size) 覆蓋。
- 關鍵字引數:
root (Path or str, optional) –
GenDGRLExperienceReplay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,則預設為 ~/.cache/torchrl/atari。download (bool 或 str, optional) – 如果找不到資料集,是否應下載。預設為
True。下載也可以傳遞為"force",在這種情況下,下載的資料將被覆蓋。sampler (Sampler, optional) – 要使用的取樣器。如果未提供,將使用預設的 RandomSampler()。
writer (Writer, optional) – 要使用的 writer。如果未提供,將使用預設的 RoundRobinWriter()。
collate_fn (callable, 可選) – 將樣本列表合併以形成 Tensor(s)/輸出的 mini-batch。在從 map 風格的資料集進行批處理載入時使用。
pin_memory (bool) – 是否應對 rb 樣本呼叫 pin_memory()。
prefetch (int, 可選) – 使用多執行緒預取的下一個批次數。
transform (Transform, optional) – 呼叫 sample() 時要執行的變換。要連結變換,請使用
Compose類。
- 變數:
available_datasets – 要下載的已接受條目的列表。這些名稱對應於 huggingface 資料集儲存庫中的目錄路徑。如果可能,該列表將從 huggingface 動態檢索。如果沒有網際網路連線,則會使用快取版本。
示例
>>> import torch >>> torch.manual_seed(0) >>> from torchrl.data.datasets import GenDGRLExperienceReplay >>> d = GenDGRLExperienceReplay("bigfish-1M_E", batch_size=32) >>> for batch in d: ... break >>> print(batch)
- add(data: TensorDictBase) int¶
將單個元素新增到重放緩衝區。
- 引數:
data (Any) – 要新增到重放緩衝區的資料
- 返回:
資料在重放緩衝區中的索引。
- append_transform(transform: Transform, *, invert: bool = False) ReplayBuffer¶
將變換附加到末尾。
呼叫 sample 時按順序應用變換。
- 引數:
transform (Transform) – 要附加的變換
- 關鍵字引數:
invert (bool, optional) – 如果為
True,則轉換將被反轉(寫入時呼叫正向呼叫,讀取時呼叫反向呼叫)。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()
- classmethod as_remote(remote_config=None)¶
建立一個遠端 ray 類的例項。
- 引數:
cls (Python Class) – 要遠端例項化的類。
remote_config (dict) – 為該類保留的 CPU 核心數量。預設為 torchrl.collectors.distributed.ray.DEFAULT_REMOTE_CLASS_CONFIG。
- 返回:
一個建立 ray 遠端類例項的函式。
- property batch_size¶
重放緩衝區的批次大小。
可以在
sample()方法中設定 batch_size 引數來覆蓋批次大小。它定義了
sample()返回的樣本數量,以及ReplayBuffer迭代器生成的樣本數量。
- property data_path¶
資料集路徑,包括分割。
- property data_path_root¶
資料集根目錄路徑。
- delete()¶
從磁碟刪除資料集儲存。
- dumps(path)¶
將重放緩衝區儲存到指定路徑的磁碟上。
- 引數:
path (Path 或 str) – 儲存重放緩衝區的路徑。
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(empty_write_count: bool = True)¶
清空重放緩衝區並將遊標重置為 0。
- 引數:
empty_write_count (bool, optional) – 是否清空 write_count 屬性。預設為 True。
- extend(tensordicts: TensorDictBase, *, update_priority: bool | None = None) torch.Tensor¶
使用資料批次擴充套件重放緩衝區。
- 引數:
tensordicts (TensorDictBase) – 用於擴充套件重放緩衝區的資料。
- 關鍵字引數:
update_priority (bool, optional) – 是否更新資料的優先順序。預設為 True。
- 返回:
已新增到重放緩衝區的資料的索引。
- insert_transform(index: int, transform: Transform, *, invert: bool = False) ReplayBuffer¶
插入變換。
呼叫 sample 時按順序執行變換。
- 引數:
index (int) – 插入變換的位置。
transform (Transform) – 要附加的變換
- 關鍵字引數:
invert (bool, optional) – 如果為
True,則轉換將被反轉(寫入時呼叫正向呼叫,讀取時呼叫反向呼叫)。預設為False。
- loads(path)¶
在給定路徑載入重放緩衝區狀態。
緩衝區應具有匹配的元件,並使用
dumps()進行儲存。- 引數:
path (Path 或 str) – 重放緩衝區儲存的路徑。
有關更多資訊,請參閱
dumps()。
- next()¶
返回重放緩衝區的下一個項。
此方法用於在 __iter__ 不可用的情況下迭代重放緩衝區,例如
RayReplayBuffer。
- 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一個包含格式化資料的新儲存。
資料轉換必須是單位化的(作用於資料集的單個樣本)。
Args 和 Keyword Args 會轉發給
map()。之後可以使用
delete()刪除資料集。- 關鍵字引數:
dest (path 或 等價物) – 新資料集位置的路徑。
num_frames (int, 可選) – 如果提供,則僅轉換前 num_frames 幀。這對於除錯轉換很有用。
返回:將在
ReplayBuffer例項中使用的新的儲存。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
鉤子目前不會在儲存重放緩衝區時序列化:每次建立緩衝區時都必須手動重新初始化它們。
- register_save_hook(hook: Callable[[Any], Any])¶
為儲存註冊儲存鉤子。
注意
鉤子目前不會在儲存重放緩衝區時序列化:每次建立緩衝區時都必須手動重新初始化它們。
- sample(batch_size: int | None = None, return_info: bool = False, include_info: bool | None = None) TensorDictBase¶
從重放緩衝區中取樣資料批次。
使用 Sampler 取樣索引,並從 Storage 中檢索它們。
- 引數:
batch_size (int, optional) – 要收集的資料的大小。如果未提供,此方法將取樣由取樣器指示的批次大小。
return_info (bool) – 是否返回資訊。如果為 True,則結果為元組 (data, info)。如果為 False,則結果為資料。
- 返回:
一個包含在重放緩衝區中選擇的資料批次的 tensordict。如果 return_info 標誌設定為 True,則包含此 tensordict 和資訊的元組。
- set_storage(storage: Storage, collate_fn: Callable | None = None)¶
在重放緩衝區中設定新的儲存並返回之前的儲存。
- 引數:
storage (Storage) – 緩衝區的新的儲存。
collate_fn (callable, optional) – 如果提供,collate_fn 將設定為此值。否則,它將被重置為預設值。
- property write_count: int¶
透過 add 和 extend 寫入緩衝區的總項數。