eval_record-test

该数据集使用LeRobot创建，包含机器人操作相关的数据。数据集结构包括2个episodes，2957帧，1个任务，4个视频。数据以parquet格式存储，视频以mp4格式存储。数据集包含动作、观测状态、图像（正面和侧面）、时间戳、帧索引、episode索引、索引和任务索引等特征。动作和观测状态的特征包括6个浮点数，分别对应机器人的不同关节位置。图像特征包括高度、宽度和通道数，正面图像分辨率为480x640，侧面图像分辨率为360x640。视频帧率为30fps，无音频。

This dataset was created using LeRobot and contains data related to robotic manipulation. The dataset structure consists of 2 episodes, 2957 frames, 1 task, and 4 videos. The tabular data is stored in Parquet format, while the videos are stored in MP4 format. The dataset includes features such as actions, observation states, images (front and side views), timestamps, frame indices, episode indices, indices, and task indices. The features for actions and observation states consist of 6 floating-point numbers, which respectively correspond to the different joint positions of the robot. The image features include height, width, and number of channels. The resolution of the front images is 480x640, while that of the side images is 360x640. The video frame rate is 30 fps, and there is no audio.