g1_05_23 (TsFile)

Apache TsFile version of Jiangeng/G1_05_23.

Overview

This dataset is a LeRobot v2.1 dataset of a Unitree G1 humanoid recorded at 50 fps, with whole-body teleoperation state, WBC action, motion tokens and SMPL body/hand poses.

• Robot: Unitree G1 humanoid
• Scale: 33 episodes, 37,420 frames, 50 fps
• Split: train
• Cameras (not uploaded): ego_view

Schema (TsFile structure)

The TsFile table is named g1_05_23.

Role	Columns
Time	Time, INT64 milliseconds
TAG	episode_index, task_index
FIELD	frame_index, sample_index
FIELD	observation_state_0 … observation_state_42
FIELD	observation_eef_state_0 … observation_eef_state_13
FIELD	action_wbc_0 … action_wbc_42
FIELD	action_motion_token_0 … action_motion_token_63
FIELD	observation_root_orientation_*
FIELD	observation_projected_gravity_*
FIELD	observation_cpp_rotation_offset_*
FIELD	observation_init_base_quat_*
FIELD	teleop_smpl_joints_* / teleop_smpl_pose_* / teleop_body_quat_w_* / teleop_target_body_orientation_*
FIELD	teleop_left_hand_joints_* / teleop_right_hand_joints_* / teleop_left_wrist_joints_* / teleop_right_wrist_joints_*
FIELD	teleop_planner_movement_* / teleop_planner_facing_* / teleop_vr_3pt_position_* / teleop_vr_3pt_orientation_*
FIELD	teleop_delta_heading, teleop_planner_speed, teleop_planner_height
FIELD	teleop_smpl_frame_index, teleop_stream_mode, teleop_planner_mode

Vector columns are flattened into scalar FLOAT measurements (single precision). The source index column is retained as sample_index.

Conversion notes

• Time = round(timestamp * 1000) with millisecond precision; time restarts inside each episode, while episode_index and task_index identify the TsFile device.
• The original timestamp field is omitted because it is exactly represented by Time / 1000.
• Camera video streams are NOT included in this repository; they remain in the source videos tree.
• The source stores floating-point vectors as double; they are flattened to single-precision FLOAT in the TsFile.
• meta/ is mirrored from the source. Aside from the redundant timestamp column and the excluded videos, no source rows or numeric fields are dropped.

Read example

from tsfile import TsFileReader

with TsFileReader("data/g1_05_23.tsfile") as reader:
    print(reader.get_all_table_schemas()["g1_05_23"])

Source & license

• Original dataset: https://huggingface.co/datasets/Jiangeng/G1_05_23
• Author: Jiangeng
• License: not declared by the original dataset; please defer to the original.