HiloHilo/eval_makermods_pick-cup

该数据集是一个机器人控制数据集，通过LeRobot创建，用于记录机器人执行任务时的动作和观察数据。数据集基于机器人类型so101_follower，包含1个完整的episode、1475帧和1个任务。数据以parquet文件格式存储，视频文件为mp4格式。数据集的特征包括：动作（action），由6个浮点数表示机器人的关节位置（如肩部平移、肩部提升、肘部弯曲、手腕弯曲、手腕滚动和夹爪位置）；观察状态（observation.state），与动作相同，提供机器人的当前状态；两个摄像头图像观察（observation.images.hand_cam和observation.images.side_cam），均为视频数据，分辨率480x640，3通道（RGB），帧率30fps，使用av1编解码器；以及元数据如时间戳、帧索引、episode索引、索引和任务索引。这些数据适用于机器人学习、控制算法开发和模拟任务。数据集结构支持训练分割（train: 0:1），总数据文件大小为100MB，视频文件大小为500MB。

This dataset is a robotics control dataset created using LeRobot, designed to record actions and observations during robot task execution. It is based on the robot type so101_follower and includes 1 complete episode, 1475 frames, and 1 task. The data is stored in parquet file format, with video files in mp4 format. The dataset features include: action, represented by 6 floating-point numbers indicating robot joint positions (e.g., shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper position); observation state (observation.state), identical to the action, providing the robots current state; two camera image observations (observation.images.hand_cam and observation.images.side_cam), both as video data with 480x640 resolution, 3 channels (RGB), 30fps frame rate, using the av1 codec; and metadata such as timestamp, frame index, episode index, index, and task index. This data is suitable for robot learning, control algorithm development, and simulation tasks. The dataset structure supports a training split (train: 0:1), with a total data file size of 100MB and video file size of 500MB.