From eecf32e77a73c3e4ad243021719a41adaa3b0f82 Mon Sep 17 00:00:00 2001 From: Haskely Date: Wed, 26 Feb 2025 00:27:36 +0800 Subject: [PATCH] feat: Add root directory option for dataset configuration (#765) Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com> --- lerobot/common/datasets/factory.py | 5 ++++- lerobot/configs/default.py | 2 ++ 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/lerobot/common/datasets/factory.py b/lerobot/common/datasets/factory.py index fb1fe6d6f..38c01b42f 100644 --- a/lerobot/common/datasets/factory.py +++ b/lerobot/common/datasets/factory.py @@ -83,10 +83,13 @@ def make_dataset(cfg: TrainPipelineConfig) -> LeRobotDataset | MultiLeRobotDatas ) if isinstance(cfg.dataset.repo_id, str): - ds_meta = LeRobotDatasetMetadata(cfg.dataset.repo_id, revision=cfg.dataset.revision) + ds_meta = LeRobotDatasetMetadata( + cfg.dataset.repo_id, root=cfg.dataset.root, revision=cfg.dataset.revision + ) delta_timestamps = resolve_delta_timestamps(cfg.policy, ds_meta) dataset = LeRobotDataset( cfg.dataset.repo_id, + root=cfg.dataset.root, episodes=cfg.dataset.episodes, delta_timestamps=delta_timestamps, image_transforms=image_transforms, diff --git a/lerobot/configs/default.py b/lerobot/configs/default.py index a50134315..0dc7f4c73 100644 --- a/lerobot/configs/default.py +++ b/lerobot/configs/default.py @@ -29,6 +29,8 @@ class DatasetConfig: # "dataset_index" into the returned item. The index mapping is made according to the order in which the # datsets are provided. repo_id: str + # Root directory where the dataset will be stored (e.g. 'dataset/path'). + root: str | None = None episodes: list[int] | None = None image_transforms: ImageTransformsConfig = field(default_factory=ImageTransformsConfig) revision: str | None = None