record2-test

该数据集是一个机器人控制数据集，使用LeRobot创建。数据集包含20个episodes，总计11825帧，涉及1个任务。数据以parquet文件存储，总数据文件大小为100MB，视频文件大小为500MB，帧率为30fps。数据集包含动作、观测状态、图像观测（来自front和pc_cam两个视角）、时间戳、帧索引、episode索引、索引和任务索引等特征。动作和观测状态均包含6个关节位置：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。图像观测为视频格式，分辨率为480x640，3通道，使用av1编码，无音频。数据集适用于机器人学习任务，如模仿学习或强化学习。This is a robotic control dataset created using LeRobot. It contains 20 episodes, totaling 11825 frames, and involves 1 single task. The dataset is stored as Parquet files, with a total data file size of 100 MB, while the video files have a size of 500 MB and a frame rate of 30 fps. The dataset includes features such as actions, observation states, image observations from two viewpoints (front and pc_cam), timestamps, frame indices, episode indices, indices, and task indices. Both actions and observation states contain 6 joint positions: shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position. The image observations are in video format, with a resolution of 480×640, 3 channels, encoded with AV1, and have no audio. This dataset is applicable to robotic learning tasks such as imitation learning or reinforcement learning.