stevenworkspace/eval_take_act_3
收藏资源简介:
该数据集是一个机器人学数据集,使用LeRobot创建,专为移动AI机器人设计。数据集包含一个总任务(total_tasks: 1),一个episode(total_episodes: 1),总计1200帧(total_frames: 1200),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。特征包括动作(action)和观测状态(observation.state),各由16个浮点数值组成,代表左右机械臂的关节位置(如left_joint_0.pos到right_joint_5.pos)以及速度和角速度(x.vel, theta.vel)。观测图像来自三个摄像头:cam_high、cam_left_wrist和cam_right_wrist,每个视频分辨率为480x640,3通道,使用AV1编解码器。此外,数据集包含时间戳、帧索引、episode索引等元数据。数据分割为训练集(splits: train: 0:1),适用于机器人控制和视觉任务研究。许可证为Apache 2.0。
This dataset is a robotics dataset created using LeRobot, designed for mobile AI robots. It includes one total task (total_tasks: 1), one episode (total_episodes: 1), with a total of 1200 frames (total_frames: 1200) at 30 fps. The data is stored in parquet format, with a total data file size of 100 MB and video file size of 200 MB. Features include action and observation.state, each consisting of 16 float32 values representing joint positions of left and right robotic arms (e.g., left_joint_0.pos to right_joint_5.pos) as well as linear and angular velocities (x.vel, theta.vel). Observation images are captured from three cameras: cam_high, cam_left_wrist, and cam_right_wrist, each video with a resolution of 480x640, 3 channels, using AV1 codec. Additionally, the dataset contains metadata such as timestamp, frame index, episode index, etc. The data is split into a training set (splits: train: 0:1), suitable for research in robot control and vision tasks. The license is Apache 2.0.




