rollout_recap_stackblocks_iter0_9_20260704_193631

该数据集是一个专为机器人学习设计的堆叠积木任务数据集，包含2个总episode、2387个总帧和1个总任务，帧率为30 fps。数据集特征包括动作（以6个浮点数表示机器人关节位置）、观测状态（与动作相同的6个浮点数）、观测图像（来自顶部和手腕摄像头的视频，分辨率为480x640，3通道，采用av1编码）、干预（布尔值表示是否发生干预）、下一个成功标志（布尔值表示任务是否成功）、时间戳、帧索引、episode索引、索引和任务索引。数据以parquet文件格式存储，视频以mp4格式存储，总数据大小为100 MB，视频大小为200 MB。机器人类型为so_follower，支持训练分割。

This dataset is designed for robot learning, specifically for block stacking tasks. It includes 2 total episodes, 2387 total frames, and 1 total task, with a frame rate of 30 fps. The dataset features include actions (represented by 6 floating-point numbers for robot joint positions), observation states (the same 6 floating-point numbers as actions), observation images (videos from top and wrist cameras with a resolution of 480x640, 3 channels, encoded in av1), intervention (a boolean indicating whether intervention occurred), next success flag (a boolean indicating task success), timestamp, frame index, episode index, index, and task index. Data is stored in parquet file format, videos in mp4 format, with a total data size of 100 MB and video size of 200 MB. The robot type is so_follower, and it supports training splits.