record-test

该数据集是一个机器人控制数据集，使用LeRobot框架创建，包含5个完整episodes和总计2215帧数据。数据集特征包括：6维浮点动作向量（控制肩部平移、肩部抬升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置）、6维浮点状态观测向量（对应关节位置）、前端摄像头视频观测（480x640分辨率RGB图像，30fps，AV1编码）、时间戳、帧索引、episode索引和任务索引。机器人类型为so_follower，数据以parquet格式存储，总数据文件大小约100MB，视频文件大小约200MB。数据集已划分为训练集（包含所有5个episodes），适用于机器人学习和控制任务。This dataset is a robot control dataset created using the LeRobot framework, containing 5 complete episodes and a total of 2215 frames. Dataset features include: 6-dimensional floating-point action vectors (controlling shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position), 6-dimensional floating-point state observation vectors (corresponding to joint positions), front camera video observations (480x640 resolution RGB images, 30fps, AV1 encoding), timestamps, frame indices, episode indices, and task indices. The robot type is so_follower, data is stored in parquet format, with a total data file size of approximately 100MB and video file size of approximately 200MB. The dataset has been divided into a training set (including all 5 episodes) and is suitable for robot learning and control tasks.