so100_pick_cube_new23

该数据集是一个机器人技术相关的数据集，使用LeRobot创建。包含5个episodes、995帧数据，主要用于训练任务。数据集结构包含动作（6维浮点数组）、奖励（1维浮点数）、完成标志（布尔值）、离散惩罚（1维浮点数）、顶部摄像头图像（3通道128x128视频）、状态（18维浮点数组）等多种特征。视频数据以10fps的帧率存储，总数据文件大小为100MB，视频文件大小为200MB。训练集包含所有5个episodes的数据。

This is a robotics-related dataset developed using the LeRobot framework. It contains 5 episodes and 995 frames of data, and is primarily intended for training tasks. The dataset includes various features: actions (6-dimensional floating-point arrays), rewards (1-dimensional floating-point numbers), termination flags (boolean values), discrete penalties (1-dimensional floating-point numbers), top-view camera images (3-channel 128×128 videos), states (18-dimensional floating-point arrays), and other related features. The video data is stored at a frame rate of 10 fps. The non-video data files of the dataset have a total size of 100 MB, while the video files occupy a total size of 200 MB. The training set includes data from all 5 episodes.