record-test_20260605_180525
收藏资源简介:
该数据集使用LeRobot工具创建,专注于机器人学任务。它包含1个总集、584个总帧和1个总任务,数据以每秒30帧的速率采集。特征包括动作(如肩部、肘部、手腕和夹爪的位置)、观察状态(类似位置信息)以及来自前置和侧置摄像头的图像观察(分辨率为480x640,3通道)。数据以parquet文件格式存储,视频文件使用AV1编码。机器人类型为follower,数据集结构还包括时间戳、帧索引、集索引等元数据。许可证为Apache 2.0。
This dataset is developed using the LeRobot toolkit and focuses on robotics-related tasks. It includes 1 overall dataset split, 584 total frames, and 1 unified task, with data collected at a frame rate of 30 frames per second (fps). The dataset features cover robot actions (e.g., positions of the shoulder, elbow, wrist, and gripper), observation states (analogous to positional information), and image observations captured by the front and side-mounted cameras, which have a resolution of 480×640 and 3 color channels. The data is stored in Parquet file format, while the video files employ the AV1 encoding standard. The robot type is follower, and the dataset structure also contains metadata such as timestamps, frame indices, and split indices. The license for this dataset is Apache 2.0.
数据集概览
数据集名称:record-test_20260605_180525
许可证:Apache-2.0
任务类型:机器人学(Robotics)
标签:LeRobot
数据构成
- 总片段数:1
- 总帧数:584
- 总任务数:1
- 帧率(FPS):30
- 数据文件大小:约100 MB
- 视频文件大小:约200 MB
数据集结构
数据特征:
| 特征名称 | 数据类型 | 形状 | 描述 |
|---|---|---|---|
action |
float32 | [6] | 动作数据,包含 shoulder_pan、shoulder_lift、elbow_flex、wrist_flex、wrist_roll、gripper 的位置信息 |
observation.state |
float32 | [6] | 状态数据,与动作数据维度相同,同样包含6个关节的位置信息 |
observation.images.front |
video | [480, 640, 3] | 前视摄像头图像,分辨率480x640,3通道,AV1编码,30fps |
observation.images.side |
video | [480, 640, 3] | 侧视摄像头图像,分辨率480x640,3通道,AV1编码,30fps |
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 |
机器人类型:so_follower
数据集划分:
- 训练集:全部数据(0至1)
文件路径结构
- 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
技术细节
- 代码版本:v3.0
- 视频编码格式:AV1
- 视频像素格式:yuv420p
- 视频后端:pyav
引用信息
目前未提供引用信息,需要更多信息。



