Dataset Viewer
Auto-converted to Parquet Duplicate
codebase_version
string
robot_type
string
total_episodes
int64
total_frames
int64
total_tasks
int64
total_videos
int64
total_chunks
int64
chunks_size
int64
fps
int64
splits
dict
data_path
string
features
dict
video_path_original
string
tsfile_conversion
dict
v2.1
reachy2
3
14,983
1
50
1
1,000
30
{ "train": "0:50" }
data/v21_apple_storage.tsfile
{ "Time": { "dtype": "int64", "shape": [ 1 ], "tsfile_role": "TIME", "unit": "ms" }, "episode_index": { "dtype": "int64", "shape": [ 1 ], "tsfile_role": "TAG" }, "task_index": { "dtype": "int64", "shape": [ 1 ], "tsfile_role": "TAG" }, ...
videos/chunk-{episode_chunk:03d}/{video_key}/episode_{episode_index:06d}.mp4
{ "source_dataset": "rerun/v21_apple_storage", "source_data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet", "converted_data_path": "data/v21_apple_storage.tsfile", "table_name": "v21_apple_storage", "granularity": "merged", "time_precision": "ms", "time_mapping": { "source...

Apple Storage v2.1 (TsFile)

Apache TsFile version of rerun/v21_apple_storage.

Overview

A small LeRobot (v2.1) sample dataset of an apple-storage manipulation task on a Reachy 2 humanoid robot, recorded at 30 fps. It contains the first three episodes of the pollen-robotics/apple_storage dataset and is used to test the LeRobot dataloader in Rerun. Each episode is a trajectory for the single task "place the apple in the plate", with synchronized joint state and action commands.

  • Robot: reachy2, 19-DoF.
  • Episodes: 3 (episode_index 0–2; lengths 299 / 300 / 300).
  • Frames: 899 time-series rows.
  • Task: 1 — place the apple in the plate.
  • Sampling rate: 30 fps (from meta/info.json).

Note: the source meta/info.json records total_frames: 14983 (a value carried over from the full upstream dataset). The actual frame data in this sample is 899 rows, matching meta/episodes.jsonl (299 + 300 + 300).

Schema (TsFile structure)

All 3 episodes are stored in a single TsFile table; episode_index and task_index are TAG columns (the TsFile device dimension), so one episode is selected with WHERE episode_index=0.

  • Time (INT64, milliseconds) — round(timestamp * 1000); restarts at 0 for each episode, stepping by ~33 ms (30 fps).
  • episode_index (TAG) — source episode id, 0–2.
  • task_index (TAG) — source task id (0).
  • frame_index (FIELD, INT64) — frame number within the episode.
  • sample_index (FIELD, INT64) — the source global index column, renamed.
  • observation_state_0 … _18 (FIELD, FLOAT) — robot joint state, flattened from the 19-element observation.state vector.
  • action_0 … _18 (FIELD, FLOAT) — joint action command, flattened from the 19-element action vector.

The source timestamp column is dropped because it equals Time / 1000 seconds. No other columns or rows are dropped.

Videos

The camera video stream of the original dataset (observation.image) is NOT included in this repository. Obtain it from the original dataset: https://huggingface.co/datasets/rerun/v21_apple_storage (the videos/ directory).

Usage

Read the .tsfile file with the Apache TsFile Java or Python SDK.

Source & license

Downloads last month
53