Hugging Face's logo

Datasets:
THULab
/
grab_dibble

Tasks:
Robotics
Modalities:
Time-series
Size:
1K<n<10K
Tags:
tsfile
timeseries
time-series
LeRobot
robotics
format:tsfile
License:
Dataset card Files
xet
 Community
grab_dibble / README.md
wangdx25's picture
wangdx25
Add TsFile (converted from ymatari/grab-dibble)
6d8dd74 verified
preview code
|
Raw
History Blame Contribute Delete
2.83 kB
metadata 
license: apache-2.0
task_categories:
  - robotics
tags:
  - tsfile
  - timeseries
  - time-series
  - LeRobot
  - robotics
  - format:tsfile
pretty_name: grab_dibble
size_categories:
  - 1K<n<10K

grab_dibble (TsFile)

Apache TsFile version of ymatari/grab-dibble.

Overview

A LeRobot v2.1 single-arm manipulation dataset recorded on a SO-101 follower robot. The arm grasps a dibble and places it into a container. Each frame holds the 6-DOF joint positions (shoulder pan/lift, elbow flex, wrist flex/roll, gripper) for both the commanded action and the observed observation.state, plus three camera views (top, surface, wrist) that are stored as videos in the original dataset.

  • Episodes: 50
  • Frames: 7,642
  • Sampling rate: 30 fps
  • Tasks: 1 — "place the dibble in the container"

Schema (TsFile structure)

All episodes share one TsFile with episode_index and task_index as TAG columns; query a single episode with WHERE episode_index = N.

  • Time (INT64, milliseconds) — round(timestamp * 1000); the source timestamp column is dropped (it equals Time / 1000).
  • episode_index (TAG) — episode identifier.
  • task_index (TAG) — task identifier (single task here).
  • frame_index (INT64) — per-episode frame counter.
  • sample_index (INT64) — source index column, renamed.
  • action_0..action_5 (FLOAT) — commanded joint positions.
  • observation_state_0..observation_state_5 (FLOAT) — observed joint positions.

The 6 action/state dimensions map to shoulder_pan.pos, shoulder_lift.pos, elbow_flex.pos, wrist_flex.pos, wrist_roll.pos, gripper.pos.

Usage

Install the Apache TsFile Python SDK (pip install tsfile) and read a converted file:

from pathlib import Path
from tsfile import TsFileReader

path = Path("data/grab_dibble.tsfile")
with TsFileReader(str(path)) as reader:
    schemas = reader.get_all_table_schemas()
    print("tables:", list(schemas))
    table_name = next(iter(schemas))
    table = schemas[table_name]
    columns = [column.get_column_name() for column in table.get_columns()]
    print("columns:", columns)
    field_names = [
        column.get_column_name()
        for column in table.get_columns()
        if column.get_column_name() not in {"Time", "time"}
    ]
    if field_names:
        with reader.query_table(table_name, field_names[:3], batch_size=1024) as result:
            batch = result.read_arrow_batch()
            if batch is not None:
                print(batch.to_pandas().head())

Source & license