eval_debug_task_cell_8_12h26m_25-jun-2026
收藏Hugging Face2026-06-25 更新2026-06-25 收录
下载链接:
https://huggingface.co/datasets/villekuosmanen/eval_debug_task_cell_8_12h26m_25-jun-2026
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是一个用于机器人任务的数据集,由LeRobot创建。数据集包含1个episode、1064个frames和1个任务,总数据文件大小为100MB,视频文件大小为200MB,帧率为20fps。数据以parquet格式存储,并分为训练集(splits: train)。特征包括:动作(action),包含12个浮点数表示的机器人关节位置(如左/右肩、肘、腕和夹爪的位置);观测状态(observation.state),同样包含12个浮点数表示的关节位置;观测图像(observation.images),包括来自左腕摄像头、右腕摄像头和顶部摄像头的视频数据,分辨率分别为720x1280和576x1024,视频编码为av1,无音频;以及时间戳、帧索引、episode索引、索引和任务索引等元数据字段。数据集适用于机器人控制、视觉导航等研究任务。
This dataset is designed for robotic tasks and created by LeRobot. It contains 1 episode, 1064 frames, and 1 task, with a total data file size of 100MB and video file size of 200MB, at a frame rate of 20fps. The data is stored in parquet format and divided into a training set (splits: train). Features include: action, consisting of 12 floating-point numbers representing robot joint positions (e.g., positions of left/right shoulder, elbow, wrist, and gripper); observation.state, also containing 12 floating-point numbers for joint positions; observation.images, which includes video data from left wrist camera, right wrist camera, and top camera, with resolutions of 720x1280 and 576x1024, video encoding as av1, and no audio; as well as metadata fields such as timestamp, frame index, episode index, index, and task index. The dataset is suitable for research tasks in robot control, visual navigation, and similar areas.
提供机构:
villekuosmanen创建时间:
2026-06-25
原始信息汇总
数据集概述
- 数据集名称:
eval_debug_task_cell_8_12h26m_25-jun-2026 - 许可证: Apache-2.0
- 任务类别: 机器人学 (Robotics)
- 标签: LeRobot
数据集结构
该数据集使用 LeRobot 创建,包含以下核心信息:
元数据 (meta/info.json)
| 属性 | 值 |
|---|---|
| 代码库版本 | v3.0 |
| 机器人类型 | so-101 |
| 总片段数 (episodes) | 1 |
| 总帧数 (frames) | 1064 |
| 总任务数 (tasks) | 1 |
| 块大小 (chunks_size) | 1000 |
| 数据文件大小 (MB) | 100 |
| 视频文件大小 (MB) | 200 |
| 帧率 (fps) | 20 |
| 数据路径 | data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet |
| 视频路径 | videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4 |
| 划分方式 | 训练集: 0:1 |
特征 (Features)
动作 (action)
- 数据类型: float32
- 形状: [12]
- 维度名称:
- left_shoulder_pan.pos
- left_shoulder_lift.pos
- left_elbow_flex.pos
- left_wrist_flex.pos
- left_wrist_roll.pos
- left_gripper.pos
- right_shoulder_pan.pos
- right_shoulder_lift.pos
- right_elbow_flex.pos
- right_wrist_flex.pos
- right_wrist_roll.pos
- right_gripper.pos
观测状态 (observation.state)
- 数据类型: float32
- 形状: [12]
- 维度名称: 与动作 (action) 特征相同
观测图像 (observation.images)
| 图像名称 | 数据类型 | 形状 (H x W x C) | 编码格式 | 帧率 (fps) | 分辨率 |
|---|---|---|---|---|---|
| left_wrist | video | 720 x 1280 x 3 | av1 (yuv420p) | 20 | 720p |
| right_wrist | video | 720 x 1280 x 3 | av1 (yuv420p) | 20 | 720p |
| top | video | 576 x 1024 x 3 | av1 (yuv420p) | 20 | 576p |
其他元数据特征
- timestamp: float32, 形状 [1]
- frame_index: int64, 形状 [1]
- episode_index: int64, 形状 [1]
- index: int64, 形状 [1]
- task_index: int64, 形状 [1]
引用信息
BibTeX 引用信息暂缺 ([More Information Needed])



