eval_two-cam-record-test
收藏Hugging Face2026-05-31 更新2026-05-31 收录
下载链接:
https://huggingface.co/datasets/alimerido/eval_two-cam-record-test
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是使用LeRobot创建的机器人学数据集,专门用于机器人控制和学习任务。数据集包含多个特征,如动作(action)和观测状态(observation.state),其中动作和状态都涉及六个关节位置(包括肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置)。观测部分还包括来自前视(front)和侧视(side)摄像头的图像,每张图像的分辨率为480x640像素,具有3个颜色通道。数据集以parquet文件格式存储,视频文件以mp4格式存储,帧率为30fps。机器人类型为so_follower,数据被分成多个块(chunks),每个块大小为1000帧。数据总大小包括100MB的数据文件和200MB的视频文件。数据集结构还包括时间戳、帧索引、episode索引、索引和任务索引等元数据,以支持机器人任务的分析和训练。
This is a robotics dataset developed using LeRobot, specifically designed for robotic control and learning tasks. It contains multiple features such as action and observation.state, both of which involve six joint positions: shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position. The observation section also includes images from front-facing and side-facing cameras, each with a resolution of 480x640 pixels and 3 color channels. The dataset is stored in Parquet file format, while the video files are in MP4 format with a frame rate of 30 fps. The robotic platform type is so_follower. The data is divided into multiple chunks, with each chunk containing 1000 frames. The total size of the dataset includes 100 MB of data files and 200 MB of video files. The dataset structure also includes metadata such as timestamps, frame indices, episode indices, indexes, and task indices to support robotic task analysis and training.
提供机构:
alimerido创建时间:
2026-05-31
原始信息汇总
数据集概述
- 名称: eval_two-cam-record-test
- 许可证: Apache-2.0
- 任务类别: 机器人学 (Robotics)
- 标签: LeRobot
数据集详情
- 数据大小: 数据文件约 100 MB,视频文件约 200 MB
- 总片段数: 6
- 总帧数: 4746
- 总任务数: 1
- 帧率: 15 FPS
- 机器人类型: so_follower
- 数据路径格式:
data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet - 视频路径格式:
videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4 - 数据分割: 训练集 (片段 0 至 5)
特征结构
- action: 6维 float32 数组,包含 shoulder_pan.pos, shoulder_lift.pos, elbow_flex.pos, wrist_flex.pos, wrist_roll.pos, gripper.pos
- observation.state: 6维 float32 数组,与 action 具有相同的命名维度
- observation.images.front: 视频数据,分辨率 480x640,3通道,H.264编码,帧率 15 FPS
- observation.images.side: 视频数据,分辨率 480x640,3通道,H.264编码,帧率 15 FPS
- timestamp: 1维 float32 数组
- frame_index: 1维 int64 数组
- episode_index: 1维 int64 数组
- index: 1维 int64 数组
- task_index: 1维 int64 数组
数据集结构信息
- 代码库版本: v3.0
- 数据分块大小: 1000
引用信息
- BibTeX: 尚未提供



