radoolonto/angled_peg_05_21_cam_ref

该数据集使用LeRobot创建，是一个机器人学习数据集，专注于Franka机器人执行任务。数据集包含119个训练集，总计67,092帧数据，帧率为15 fps。数据特征包括：来自两个摄像头（cam0和cam1）的图像观测（3通道，分辨率240x320）、机器人状态观测（3维浮点向量）、动作数据（10维浮点向量），以及时间戳、帧索引、任务索引等元数据。数据以Parquet格式存储，总数据文件大小为100 MB，视频文件大小为200 MB。数据集适用于机器人控制、模仿学习或强化学习任务。This dataset was created using LeRobot and is a robotics learning dataset focused on Franka robot task execution. It contains 119 episodes for training, with a total of 67,092 frames at 15 fps. Features include: image observations from two cameras (cam0 and cam1) with 3 channels and 240x320 resolution, robot state observations (3-dimensional float32 vector), action data (10-dimensional float32 vector), and metadata such as timestamp, frame index, and task index. Data is stored in Parquet format, with total data file size of 100 MB and video file size of 200 MB. The dataset is suitable for robotics control, imitation learning, or reinforcement learning tasks.