CoolFace
Datasetpublic

THULab/G1_WBT_Inspire_Pickup_Pillow_MainCamOnly

G1 WBT Inspire Pickup Pillow MainCamOnly TsFile This dataset is converted from unitreerobotics/G1_WBT_Inspire_Pickup_Pillow_MainCamOnly to Apache TsFile format. Modalities: Time-series. The converted repository contains numeric robot state/action trajectories and LeRobot metadata. Camera videos are not included in this TsFile conversion; they remain in the original Hugging Face dataset under its videos/ tree. Source Dataset Original dataset:… See the full description on the dataset page: https://huggingface.co/datasets/THULab/G1_WBT_Inspire_Pickup_Pillow_MainCamOnly.

sourceHugging Faceapache-2.0updated 1mo agoView on Hugging Face
0likes59downloads
Dataset Card

G1 WBT Inspire Pickup Pillow MainCamOnly TsFile

This dataset is converted from `unitreerobotics/G1_WBT_Inspire_Pickup_Pillow_MainCamOnly` to Apache TsFile format.

Modalities: Time-series. The converted repository contains numeric robot state/action trajectories and LeRobot metadata. Camera videos are not included in this TsFile conversion; they remain in the original Hugging Face dataset under its videos/ tree.

Source Dataset

  • —Original dataset: `unitreerobotics/G1_WBT_Inspire_Pickup_Pillow_MainCamOnly`
  • —License: apache-2.0
  • —Robot type from source metadata: unitree_g1
  • —Source codebase version from metadata: v3.0
  • —Source split: train, episodes 0:609
  • —Sampling rate: 30 fps
  • —Source scale: 609 episodes, 157,060 frames, and 1 task.
  • —Source data layout: G1_WB_Dex5_Pickup_Pillow/data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet
  • —Source video layout: G1_WB_Dex5_Pickup_Pillow/videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4
  • —Source card note: created using the LeRobot framework.

Converted Data

  • —TsFile path: data/g1_wbt_inspire_pickup_pillow_maincamonly.tsfile
  • —Table name: g1_wbt_inspire_pickup_pillow_maincamonly
  • —Converted rows: 157,060
  • —Episodes: 609
  • —Tasks: 1
  • —Source frame Parquet files converted: 1
  • —Output file size: 63,886,359 bytes
  • —Time precision: milliseconds

Schema

  • —Time: integer timestamp in milliseconds, computed as round(timestamp * 1000).
  • —TAG columns: episode_index, task_index.
  • —Integer FIELD columns: frame_index, sample_index.
  • —FLOAT FIELD columns:
  • —observation_state_ee_state_0 through observation_state_ee_state_11
  • —observation_state_hand_state_0 through observation_state_hand_state_11
  • —observation_state_robot_q_current_0 through observation_state_robot_q_current_35
  • —action_ee_action_0 through action_ee_action_11
  • —action_hand_cmd_0 through action_hand_cmd_11
  • —action_robot_q_desired_0 through action_robot_q_desired_35

The source card describes observation.state.ee_state and action.ee_action as 12-dimensional end-effector states/actions, observation.state.hand_state and action.hand_cmd as 12-dimensional hand states/actions in this source, and observation.state.robot_q_current / action.robot_q_desired as 36-dimensional robot configuration vectors.

Conversion Notes

  • —Vector columns are flattened while preserving source names: . is replaced with _, and the element index is appended.
  • —The source timestamp column is dropped because it is represented by Time / 1000 seconds after conversion.
  • —The source index column is renamed to sample_index.
  • —Video features observation.images.cam_0 and observation.images.cam_1 are omitted from this repository. Use the original dataset videos at `unitreerobotics/G1_WBT_Inspire_Pickup_Pillow_MainCamOnly/videos`.
  • —Source meta/ files are mirrored, with meta/info.json updated to describe the converted TsFile path and schema.

Read Example

python
from pathlib import Path

tsfile_path = Path("data/g1_wbt_inspire_pickup_pillow_maincamonly.tsfile")
print(tsfile_path)

# Read this file with the Apache TsFile Java/Python SDK or TsFile tools.