may7_TRIMMED_first_50_frames_merged
收藏Hugging Face2026-05-12 更新2026-05-12 收录
下载链接:
https://huggingface.co/datasets/jjr1007/may7_TRIMMED_first_50_frames_merged
下载链接
链接失效反馈官方服务:
资源简介:
该数据集使用LeRobot创建,专注于机器人控制任务,具体针对so_follower机器人类型。它包含271个episodes,总计231,525帧,数据以parquet文件格式存储,并附带视频文件。数据集特征包括动作(如关节位置:shoulder_pan.pos、shoulder_lift.pos等)、观测状态(与动作相同的关节位置和前端图像,图像分辨率为1080x1920,3通道)、时间戳、帧索引、episode索引等。帧率为30fps,用于训练机器人学习和控制应用。
This dataset was developed using LeRobot, focusing on robotic control tasks, specifically targeting the so_follower robot type. It contains 271 episodes, totaling 231,525 frames, with data stored in Parquet file format alongside video files. The dataset includes actions (such as joint positions: shoulder_pan.pos, shoulder_lift.pos, etc.), observation states (identical joint positions as those in the actions plus front-facing images with a resolution of 1080x1920 and 3 channels), timestamps, frame indices, episode indices, and so on. The frame rate is 30 fps, and it is intended for training robotic learning and control applications.
提供机构:
jjr1007
创建时间:
2026-05-10
原始信息汇总
数据集概述
该数据集是一个面向机器人领域的公开数据集,基于 LeRobot 框架构建。
基本信息
- 许可证: Apache-2.0
- 任务类型: 机器人学(Robotics)
- 标签: LeRobot
- 代码库版本: v3.0
- 机器人类型: so_follower
数据规模
- 总片段数 (Episodes): 271
- 总帧数 (Frames): 231,525
- 总任务数 (Tasks): 1
- 帧率 (FPS): 30
- 数据文件大小: 100 MB
- 视频文件大小: 200 MB
- 块大小 (Chunks): 1000
数据划分
所有数据(0:271)均用于训练集。
数据特征
| 特征名称 | 数据类型 | 形状 | 描述 |
|---|---|---|---|
action |
float32 | [6] | 机器人动作,包含6个关节位置值 (shoulder_pan, shoulder_lift, elbow_flex, wrist_flex, wrist_roll, gripper) |
observation.state |
float32 | [6] | 机器人观察到的状态,与动作结构相同 |
observation.images.front |
video | [1080, 1920, 3] | 前视相机视频,分辨率1080x1920,AV1编码,YUV420P格式,30 FPS,3通道,不含音频 |
timestamp |
float32 | [1] | 时间戳 |
frame_index |
int64 | [1] | 帧索引 |
episode_index |
int64 | [1] | 片段索引 |
index |
int64 | [1] | 索引 |
task_index |
int64 | [1] | 任务索引 |
数据存储路径
- 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



