sortletters_serialized_123

该数据集是一个机器人控制数据集，由LeRobot项目创建。它包含123个episodes，总计124675帧，涉及一个任务（total_tasks: 1）。数据集针对双机器人手臂（bi_so_follower）设计，特征包括：动作数据（12维浮点数组，对应左右手臂的关节位置，如肩部、肘部、腕部和夹爪位置）、观测状态（12维浮点数组，与动作相同的位置信息）、以及来自左腕（left_wrist）、右前（right_front）和右腕（right_wrist）摄像头的图像观测，图像分辨率为480x640，3通道，视频编码为h264，帧率为30fps。此外，还包含时间戳、帧索引、episode索引、任务索引等元数据。数据以parquet文件（用于结构化数据）和mp4视频文件存储，总数据大小约100MB，视频大小约200MB。数据集用于训练和评估机器人控制模型，特别适用于模仿学习或强化学习任务。This dataset is a robot control dataset created by the LeRobot project. It contains 123 episodes, totaling 124675 frames, and involves one task (total_tasks: 1). The dataset is designed for dual robot arms (bi_so_follower), with features including: action data (12-dimensional floating-point arrays corresponding to joint positions of the left and right arms, such as shoulder, elbow, wrist, and gripper positions), observation states (12-dimensional floating-point arrays with the same position information as actions), and image observations from left wrist (left_wrist), right front (right_front), and right wrist (right_wrist) cameras, with image resolution of 480x640, 3 channels, video encoding as h264, and frame rate of 30fps. Additionally, it includes metadata such as timestamps, frame indices, episode indices, and task indices. The data is stored in parquet files (for structured data) and mp4 video files, with a total data size of approximately 100MB and video size of about 200MB. The dataset is used for training and evaluating robot control models, particularly suitable for imitation learning or reinforcement learning tasks.