dogeum/eval_so101_5000

该数据集是一个用于机器人任务的数据集，使用LeRobot创建。数据集包含1个剧集、824帧数据，涉及1个任务。数据以parquet文件格式存储，总大小为100MB，视频文件大小为500MB，帧率为30fps。数据集的特征包括动作（如肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置）、观测状态（与动作相同的关节位置）、来自前置和顶部摄像头的图像观测（分辨率为480x640，3通道彩色视频），以及时间戳、帧索引、剧集索引、索引和任务索引等元数据。视频使用AV1编码，像素格式为yuv420p，无音频。数据集适用于机器人控制、模仿学习或强化学习等研究。This dataset is designed for robotics tasks and was created using LeRobot. It contains 1 episode, 824 frames, and involves 1 task. The data is stored in parquet format, with a total size of 100MB for data files and 500MB for video files, at a frame rate of 30fps. Features include actions (e.g., shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper positions), observation states (same joint positions as actions), image observations from front and top cameras (resolution 480x640, 3-channel color video), and metadata such as timestamp, frame index, episode index, index, and task index. Videos are encoded with AV1 codec, pixel format yuv420p, and no audio. The dataset is suitable for research in robotics control, imitation learning, or reinforcement learning.