rq2-push-red-block-100-hsv-sam-normalized
收藏Hugging Face2026-06-08 更新2026-06-08 收录
下载链接:
https://huggingface.co/datasets/martonbodo/rq2-push-red-block-100-hsv-sam-normalized
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是通过LeRobot创建的,专用于机器人技术任务,特别是跟随任务。它包含100个训练集,总计21726帧,帧率为30 fps。数据以parquet文件格式存储,视频文件为mp4格式。机器人类型为so_follower。数据集特征包括动作(action)、观察状态(observation.state)、来自两个摄像头(eagle和side)的图像观察(observation.images.eagle和observation.images.side),以及时间戳(timestamp)、帧索引(frame_index)、集索引(episode_index)、索引(index)和任务索引(task_index)。动作和观察状态均包含6个浮点数值,分别对应机器人关节位置(如肩部平移、肩部提升、肘部弯曲、手腕弯曲、手腕旋转和夹爪位置)。图像观察为视频格式,分辨率为240x320像素,3个通道(RGB),使用av1编解码器。该数据集主要用于训练和评估机器人控制模型。
This dataset is created via LeRobot for robotics tasks. It contains 100 training episodes with a total of 21726 frames at a frame rate of 30 fps. The data is stored in parquet file format, and video files are in mp4 format. The robot type is so_follower. The dataset includes multiple features: action, observation.state, image observations from two cameras (eagle and side) as observation.images.eagle and observation.images.side, as well as timestamp, frame_index, episode_index, index, and task_index. Both action and observation.state contain 6 floating-point values corresponding to robot joint positions (e.g., shoulder translation, shoulder lift, elbow bend, wrist bend, wrist rotation, and gripper position). The image observations are in video format with a resolution of 240x320 pixels, 3 channels (RGB), using the av1 codec. The dataset is primarily used for training and evaluating robot control models, especially for following tasks.
提供机构:
martonbodo创建时间:
2026-06-08
原始信息汇总
数据集概述
- 名称: rq2-push-red-block-100-hsv-sam-normalized
- 许可协议: Apache-2.0
- 任务类别: 机器人学(Robotics)
- 创建工具: LeRobot
数据集结构
- 代码库版本: v3.0
- 机器人类型: so_follower
- 总回合数: 100
- 总帧数: 21726
- 总任务数: 1
- 分块大小: 1000
- 数据文件大小: 100 MB
- 视频文件大小: 200 MB
- 帧率 (FPS): 30
- 数据分割: 训练集 (0:100,即全部数据用于训练)
数据特征
数据集包含以下特征字段:
| 特征名 | 数据类型 | 形状 | 描述 |
|---|---|---|---|
| action | float32 | [6] | 机器人动作,包含肩旋转、肩升降、肘弯曲、腕弯曲、腕旋转和夹爪的位置 |
| observation.state | float32 | [6] | 机器人观测状态,与动作特征相同(6个关节位置) |
| observation.images.eagle | 视频 | [240, 320, 3] | 俯视摄像头图像,分辨率240x320,RGB三通道,编码为AV1,帧率30 |
| observation.images.side | 视频 | [240, 320, 3] | 侧视摄像头图像,分辨率240x320,RGB三通道,编码为AV1,帧率30 |
| timestamp | float32 | [1] | 时间戳 |
| frame_index | int64 | [1] | 帧索引 |
| episode_index | int64 | [1] | 回合索引 |
| index | int64 | [1] | 索引 |
| task_index | int64 | [1] | 任务索引 |
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
引用信息
当前引用信息未提供。



