RAPID_dummy_50 / README.md
Ziegelll's picture
Upload README.md with huggingface_hub
220d554 verified
metadata
license: mit
task_categories:
  - robotics
tags:
  - LeRobot
  - robotic-manipulation
  - pick-and-place
  - teleoperation
configs:
  - config_name: default
    data_files: data/*/*.parquet

Rapid Dummy 50 Dataset

This dataset contains teleoperated demonstrations for a pick-and-place task using a dummy_arm_inz arm with gripper.

Dataset Description

Dataset Structure

Meta Information (meta/info.json)

{
  "codebase_version": "v2.1",
  "robot_type": "dummy_arm_inz",
  "fps": 10,
  "total_episodes": 16,
  "total_frames": 2969,
  "total_tasks": 1,
  "total_videos": 32,
  "chunks_size": 1000,
  "splits": {
    "train": "0:16"
  },
  "data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
  "video_path": "videos/chunk-{episode_chunk:03d}/{video_key}/episode_{episode_index:06d}.mp4",
  "features": {
    "observation.state": {
      "dtype": "float32",
      "shape": [
        7
      ],
      "names": [
        "joint_0",
        "joint_1",
        "joint_2",
        "joint_3",
        "joint_4",
        "joint_5",
        "gripper_position_deg"
      ]
    },
    "gripper_torque": {
      "dtype": "float32",
      "shape": [
        1
      ],
      "names": [
        "gripper_torque"
      ]
    },
    "action": {
      "dtype": "float32",
      "shape": [
        7
      ],
      "names": [
        "joint_0",
        "joint_1",
        "joint_2",
        "joint_3",
        "joint_4",
        "joint_5",
        "gripper"
      ]
    },
    "episode_index": {
      "dtype": "int64",
      "shape": []
    },
    "frame_index": {
      "dtype": "int64",
      "shape": []
    },
    "timestamp": {
      "dtype": "float64",
      "shape": []
    },
    "clock_time": {
      "dtype": "float64",
      "shape": []
    },
    "next.done": {
      "dtype": "bool",
      "shape": []
    },
    "index": {
      "dtype": "int64",
      "shape": []
    },
    "task_index": {
      "dtype": "int64",
      "shape": []
    },
    "task": {
      "dtype": "string",
      "shape": []
    },
    "observation.images.cam_head": {
      "dtype": "video",
      "shape": [
        720,
        1280,
        3
      ],
      "names": [
        "height",
        "width",
        "channel"
      ],
      "info": {
        "video.fps": 10,
        "video.codec": "mp4v"
      }
    },
    "observation.images.cam_wrist": {
      "dtype": "video",
      "shape": [
        720,
        1280,
        3
      ],
      "names": [
        "height",
        "width",
        "channel"
      ],
      "info": {
        "video.fps": 10,
        "video.codec": "mp4v"
      }
    }
  }
}

Features

The dataset follows the LeRobot vv2.1 format with the following features:

Feature Type Shape Description
observation.state float32 [7] observation.state data
gripper_torque float32 [1] Gripper torque feedback
action float32 [7] Action commands (1 dimensions)
episode_index int64 [] Episode identifier
frame_index int64 [] Frame number within episode
timestamp float64 [] Timestamp in seconds
clock_time float64 [] clock_time data
next.done bool [] Episode termination flag
index int64 [] index data
task_index int64 [] task_index data
task string [] Task description
observation.images.cam_head video [720, 1280, 3] Cam_head camera RGB video
observation.images.cam_wrist video [720, 1280, 3] Cam_wrist camera RGB video

Data Format

The dataset follows the LeRobot format:

dataset/
├── data/
│   └── chunk-000/
│       └── episode_*.parquet
├── videos/
│   └── chunk-000/
│       ├── observation.images.cam_head/
│       │   └── episode_*.mp4
│       └── observation.images.cam_wrist/
│           └── episode_*.mp4
└── meta/
    ├── info.json
    ├── episodes.jsonl
    ├── episodes_stats.jsonl
    └── tasks.jsonl

Usage

Load the dataset using LeRobot:

from lerobot.common.datasets.lerobot_dataset import LeRobotDataset

dataset = LeRobotDataset("Ziegelll/RAPID_dummy_50")

Citation

If you use this dataset, please cite:

@dataset{Ziegelll_RAPID_dummy_50_dataset_2026,
    title={Rapid Dummy 50 Dataset},
    author={Dataset Creator},
    year={2026},
    publisher={Hugging Face},
    url={https://huggingface.co/datasets/Ziegelll/RAPID_dummy_50}
}

Generated on 2026-01-24 21:58:10 with LeRobot dataset tools.