HSJUSER/ffw_sg2_rev1_Test_A2

该数据集是一个机器人数据集，使用LeRobot创建，专门用于机器人学习任务。数据集包含9个完整的情节，总计5404帧，对应27个视频，帧率为30fps。数据以Parquet格式存储，并分为训练集（涵盖所有情节）。数据集的特征包括时间戳、帧索引、情节索引、任务索引，以及观测数据：来自头部摄像头（分辨率376x672）、左腕摄像头（分辨率240x424）和右腕摄像头（分辨率240x424）的图像，以及一个19维的关节状态观测向量（涵盖左右臂、夹持器和头部关节）。动作数据同样是一个19维的关节动作向量。数据集的机器人类型为ffw_sg2_rev1，适用于机器人控制、模仿学习或强化学习等应用场景。

This dataset is a robotics dataset created using LeRobot, specifically designed for robot learning tasks. It contains 9 total episodes with 5404 frames, corresponding to 27 videos at a frame rate of 30fps. The data is stored in Parquet format and split into a training set (covering all episodes). Dataset features include timestamps, frame indices, episode indices, task indices, and observations: images from a head camera (resolution 376x672), left wrist camera (resolution 240x424), and right wrist camera (resolution 240x424), as well as a 19-dimensional joint state observation vector (covering left and right arm, gripper, and head joints). Action data is also a 19-dimensional joint action vector. The robot type is ffw_sg2_rev1, making it suitable for applications such as robot control, imitation learning, or reinforcement learning.