eval_so101-full-fold-merged-smolvla-g

该数据集是使用LeRobot创建的机器人数据集，专为机器人任务设计。它采用Apache-2.0许可证，属于机器人学任务类别。数据集结构基于元数据文件info.json，代码库版本为v3.0，机器人类型为so_follower。数据集包含动作特征（如肩部、肘部、腕部和夹持器的位置）、观测状态特征（与动作特征相同）、夹持器图像观测（以视频格式存储，分辨率为480x640x3），以及时间戳、帧索引、剧集索引、索引和任务索引等元数据。数据以parquet文件格式存储，视频以mp4格式存储，总数据文件大小为100MB，视频文件大小为200MB，帧率为30fps。数据集未提供总剧集数、总帧数和总任务数（在元数据中标记为0），这可能表示这些信息需从实际数据中获取。

This robot dataset was created using LeRobot and is specifically designed for robotics tasks. It is licensed under Apache-2.0 and falls under the robotics task category. The dataset structure is based on the metadata file info.json, with a codebase version of v3.0 and a robot type of so_follower. The dataset includes action features (e.g., positions of the shoulder, elbow, wrist, and gripper), observation state features (identical to the action features), gripper image observations stored in video format with a resolution of 480x640x3, and metadata such as timestamps, frame indices, episode indices, indices, and task indices. The structured data is stored in Parquet file format, while the videos are stored in MP4 format. The total size of the data files is 100 MB, the total size of the video files is 200 MB, and the frame rate is 30 fps. The dataset does not provide the total number of episodes, total number of frames, and total number of tasks (marked as 0 in the metadata), which may indicate that this information needs to be retrieved from the actual dataset.

原始信息汇总

数据集概述

• 数据集名称: eval_so101-full-fold-merged-smolvla-g
• 许可证: Apache-2.0
• 任务类别: 机器人学 (Robotics)
• 标签: LeRobot
• 创建工具: 使用 LeRobot 创建

数据集结构

• 机器人类型: so_follower
• 总片段数: 0
• 总帧数: 0
• 总任务数: 0
• 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

特征信息

特征名称	数据类型	形状	描述
action	float32	[6]	包含 shoulder_pan.pos, shoulder_lift.pos, elbow_flex.pos, wrist_flex.pos, wrist_roll.pos, gripper.pos
observation.state	float32	[6]	包含 shoulder_pan.pos, shoulder_lift.pos, elbow_flex.pos, wrist_flex.pos, wrist_roll.pos, gripper.pos
observation.images.gripper	video	[480, 640, 3]	高度480，宽度640，3通道视频图像
timestamp	float32	[1]	时间戳
frame_index	int64	[1]	帧索引
episode_index	int64	[1]	片段索引
index	int64	[1]	全局索引
task_index	int64	[1]	任务索引

引用信息

• 主页: 暂无信息
• 论文: 暂无信息
• 引用格式 (BibTeX): 暂无信息