eval_omx_cube_1pos_50ep_1k_260607

该数据集使用LeRobot创建，是一个专门用于OMX Follower机器人的机器人学相关数据集。数据集包含10个episodes，总计8790帧，采样率为30fps。数据以parquet文件格式存储，特征包括动作（6维浮点数组，对应机器人关节位置）、观测状态（6维浮点数组，表示机器人关节位置）、前视图像观测（480x640x3视频帧）、时间戳、帧索引和episode索引。数据集旨在用于训练和评估机器人控制任务，所有数据已划分为训练集。This dataset is created using LeRobot and is a robotics-related dataset specifically designed for the OMX Follower robot. It contains 10 episodes, totaling 8790 frames with a sampling rate of 30fps. The data is stored in Parquet file format, featuring actions (6-dimensional float arrays corresponding to robot joint positions), observation states (6-dimensional float arrays representing robot joint positions), front-facing image observations (480x640x3 video frames), timestamps, frame indices, and episode indices. The dataset is intended for training and evaluating robot control tasks, with all data partitioned into a training set.