omnaathg/so101_card_memory

该数据集是使用LeRobot工具创建的机器人学数据集，专门用于机器人控制任务。数据集包含60个总集数，45000个总帧数，以及3个总任务。数据以Parquet文件格式存储，总数据文件大小为100MB，视频文件大小为200MB。数据集的特征包括：动作（action）和观察状态（observation.state），均为浮点型数组，表示机器人关节位置（如肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部滚动、夹爪位置）；观察图像包括三脚架摄像头（tripod_cam）和夹爪摄像头（gripper_cam），均为视频格式，分辨率分别为480x640和720x1280，帧率为30fps；此外，还包括时间戳、帧索引、集索引、索引和任务索引等元数据。机器人类型为“so_follower”，数据集划分为训练集，涵盖所有集。

This dataset is a robotics dataset created using the LeRobot tool, specifically designed for robot control tasks. It contains a total of 60 episodes, 45,000 frames, and 3 tasks. The data is stored in Parquet format, with a total data file size of 100MB and video file size of 200MB. The features of the dataset include: action and observation.state, both as float32 arrays representing robot joint positions (e.g., shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, gripper position); observation images include tripod_cam and gripper_cam, both in video format with resolutions of 480x640 and 720x1280 respectively, at 30fps; additionally, it includes metadata such as timestamp, frame index, episode index, index, and task index. The robot type is so_follower, and the dataset is split into a training set covering all episodes.