skill-set-r3-train
收藏资源简介:
该数据集是一个机器人学数据集,使用LeRobot创建,包含168个episodes和122685帧数据,涉及5个不同任务。数据以parquet文件格式存储,视频文件为mp4格式。数据集特征包括动作(包含6个关节位置:肩部平移、肩部升降、肘部弯曲、腕部弯曲、腕部旋转、夹爪位置)、观察状态(相同6个关节位置)、两个相机图像(分辨率为480x640,3通道,30fps),以及时间戳、帧索引、episode索引、索引和任务索引等元数据。数据集适用于机器人控制和视觉任务研究,使用Apache 2.0许可证。
This dataset is a robotics dataset created using LeRobot, containing 168 episodes and 122,685 frames of data across 5 different tasks. The data is stored in parquet file format, with video files in mp4 format. Dataset features include actions (comprising 6 joint positions: shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position), observation states (the same 6 joint positions), two camera images (resolution 480x640, 3 channels, 30fps), as well as metadata such as timestamps, frame index, episode index, index, and task index. The dataset is suitable for research in robot control and visual tasks and is licensed under Apache 2.0.
数据集概述
- 数据集名称: skill-set-r3-train
- 许可证: Apache-2.0
- 任务类别: 机器人学 (Robotics)
- 标签: LeRobot
- 创建工具: 基于 LeRobot 创建
数据集结构
- 代码库版本: v3.0
- 机器人类型: so_follower
- 总集数 (episodes): 168
- 总帧数 (frames): 122,685
- 总任务数 (tasks): 5
- 分块大小 (chunks size): 1000
- 数据文件大小: 100 MB
- 视频文件大小: 200 MB
- 帧率 (FPS): 30
- 数据划分: 仅有训练集 (train),包含 0 到 167 共 168 个 episode
特征描述
| 特征名称 | 数据类型 | 形状 | 描述 |
|---|---|---|---|
| action | float32 | [6] | 包含 6 个关节的位置动作:shoulder_pan, shoulder_lift, elbow_flex, wrist_flex, wrist_roll, gripper |
| observation.state | float32 | [6] | 机器人状态观测,与动作维度相同 |
| observation.images.camera1 | video | [480, 640, 3] | 来自 camera1 的视频流,分辨率 480×640,3 通道,编码为 AV1,30 FPS |
| observation.images.camera2 | video | [480, 640, 3] | 来自 camera2 的视频流,参数同 camera1 |
| timestamp | float32 | [1] | 时间戳 |
| frame_index | int64 | [1] | 帧索引 |
| episode_index | int64 | [1] | 集数索引 |
| index | int64 | [1] | 全局索引 |
| task_index | int64 | [1] | 任务索引 |
数据文件路径
- 数据文件:
data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet - 视频文件:
videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4
可视化入口
- 可在 Hugging Face Space 中可视化该数据集。



