so101_d65

该数据集是通过LeRobot创建的，主要用于机器人技术领域。数据集包含60个总集数，42395个总帧数，1个总任务数，180个总视频数。数据集的结构包括多个特征，如动作、观察状态、图像（手腕、手机、笔记本电脑）以及时间戳等。动作和观察状态的特征包括6个浮点型数据，分别代表主肩旋转、主肩提升、主肘弯曲、主腕弯曲、主腕滚动和主夹持器。图像特征包括480x640x3的视频数据，帧率为30fps，编码为av1，像素格式为yuv420p。数据集的分割为训练集，包含0到60集。

This dataset was created via LeRobot, primarily intended for the field of robotics. It contains 60 total episodes, 42395 total frames, 1 total task, and 180 total videos. The dataset structure includes multiple features such as actions, observation states, images (wrist, phone, laptop), and timestamps. The features for actions and observation states are composed of 6 floating-point values, which respectively represent main shoulder rotation, main shoulder lift, main elbow flexion, main wrist flexion, main wrist roll, and main gripper. The image features are 480x640x3 video data with a frame rate of 30fps, encoded in AV1, and the pixel format is yuv420p. The dataset is split into a training set, which includes episodes 0 to 60.