fix(dataset): use revision-safe Hub cache for downloaded datasets (#3233)

* refactor(dataset): enhance dataset root directory handling and introduce hub cache support

- Updated DatasetConfig and LeRobotDatasetMetadata to clarify root directory behavior and introduce a dedicated hub cache for downloads.
- Refactored LeRobotDataset and StreamingLeRobotDataset to utilize the new hub cache and improve directory management.
- Added tests to ensure correct behavior when using the hub cache and handling different revisions without a specified root directory.

* refactor(dataset): improve root directory handling in LeRobotDataset

- Updated LeRobotDataset to store the requested root path separately from the actual root path.
- Adjusted metadata loading to use the requested root, enhancing clarity and consistency in directory management.

* refactor(dataset): minor improvements for hub cache support

* chore(datasets): guard in resume + assertion test

---------

Co-authored-by: AdilZouitine <adilzouitinegm@gmail.com>
Co-authored-by: mickaelChen <mickael.chen.levinson@gmail.com>
This commit is contained in:
Steven Palma
2026-03-27 22:21:55 +01:00
committed by GitHub
parent 975d89b38d
commit 4e45acca52
8 changed files with 440 additions and 40 deletions

View File

@@ -18,6 +18,7 @@ import importlib.resources
import json
import logging
from collections.abc import Iterator
from pathlib import Path
from typing import Any
import datasets
@@ -101,6 +102,18 @@ DEFAULT_FEATURES = {
}
def has_legacy_hub_download_metadata(root: Path) -> bool:
"""Return ``True`` when *root* looks like a legacy Hub ``local_dir`` mirror.
``snapshot_download(local_dir=...)`` stores lightweight metadata under
``<local_dir>/.cache/huggingface/download/``. The presence of this
directory is a reliable indicator that the dataset was downloaded with
the old non-revision-safe ``local_dir`` mode and should be re-fetched
through the snapshot cache instead.
"""
return (root / ".cache" / "huggingface" / "download").exists()
def update_chunk_file_indices(chunk_idx: int, file_idx: int, chunks_size: int) -> tuple[int, int]:
if file_idx == chunks_size - 1:
file_idx = 0