five

so100_block_in_cup_at_home

收藏
Hugging Face2025-06-06 更新2025-06-07 收录
下载链接:
https://huggingface.co/datasets/jccj/so100_block_in_cup_at_home
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,主要用于机器人技术领域。数据集包含48个总集数,16860个总帧数,1个总任务数,96个总视频数。数据以parquet格式存储,视频以mp4格式存储。数据集的特征包括动作、观察状态、观察图像(顶部和左腕)、时间戳、帧索引、集索引、索引和任务索引。动作和观察状态的数据类型为float32,形状为6,分别对应机器人各个关节的位置。观察图像的数据类型为视频,形状为1080x1920x3,分别对应高度、宽度和通道数。时间戳、帧索引、集索引、索引和任务索引的数据类型分别为float32和int64,形状为1。

This dataset was created using LeRobot and is primarily intended for the field of robotics. It contains 48 total episodes, 16860 total frames, 1 total task, and 96 total videos. The tabular data is stored in Parquet format, while the videos are saved in MP4 format. The features of this dataset include actions, observation states, observation images (top-view and left-wrist), timestamps, frame indices, episode indices, indices, and task indices. Both actions and observation states have a data type of float32 and a shape of 6, which correspond to the positions of each joint of the robot respectively. The observation images are video data with a shape of 1080×1920×3, corresponding to height, width, and number of channels respectively. The data types of timestamps, frame indices, episode indices, indices, and task indices are float32 and int64 respectively, each with a shape of 1.
提供机构:
jccj
创建时间:
2025-06-06
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作