thinhpt/data_test2

该数据集是一个机器人学数据集，由LeRobot创建，用于机器人控制任务。数据集包含10个完整的情节，总计3328帧数据，帧率为30fps。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。数据集仅包含训练集，涵盖一个任务。机器人类型为“so101_follower”，特征包括动作（action）和观察（observation）。动作特征为6维浮点数组，对应机器人关节位置（如肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置）。观察特征包括状态（6维关节位置）和三个摄像头图像（顶部、腕部和侧面），每个图像分辨率为480x640，3通道彩色视频，编码为av1格式。此外，数据集还包含时间戳、帧索引、情节索引、索引和任务索引等元数据。

This dataset is a robotics dataset created by LeRobot for robot control tasks. It contains 10 complete episodes with a total of 3328 frames at a frame rate of 30fps. The data is stored in parquet format, with a total data file size of 100MB and video file size of 200MB. The dataset only includes a training set and covers one task. The robot type is so101_follower, and features include actions and observations. Action features are 6-dimensional float arrays corresponding to robot joint positions (e.g., shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper position). Observation features include state (6-dimensional joint positions) and three camera images (top, wrist, and side), each with a resolution of 480x640, 3-channel color video encoded in av1 format. Additionally, the dataset includes metadata such as timestamp, frame index, episode index, index, and task index.