jogarulfop/rollout_eval_act_2026-05-29_shakeitup_dragonfly_checkpoint_20_000_20260529_144518
收藏资源简介:
该数据集是一个用于机器人学领域的数据集,使用LeRobot创建。它包含3个总片段、1868个总帧数和1个总任务。数据集的特征包括动作(6维浮点数组,对应机械臂关节位置)、观测状态(6维浮点数组,同样对应关节位置)、观测图像(包括顶部摄像头和腕部摄像头的视频数据,分辨率480x640,以及左触觉谱图视频,分辨率224x224),以及时间戳、帧索引、片段索引、索引和任务索引等元数据。数据以Parquet文件格式存储,视频文件以MP4格式存储。机器人类型为so101_follower_dragontactile,数据集总大小约为300MB(数据文件100MB,视频文件200MB)。数据集适用于训练和评估机器人控制模型,特别是涉及视觉和触觉反馈的任务。
This dataset is designed for robotics and was created using LeRobot. It includes a total of 3 episodes, 1868 frames, and 1 task. The dataset features actions (a 6-dimensional float32 array representing robotic joint positions), observation states (a similar 6-dimensional float32 array for joint positions), observation images (including video data from a top camera and a wrist camera with 480x640 resolution, and a left tactile spectrogram video with 224x224 resolution), along with metadata such as timestamps, frame indices, episode indices, indices, and task indices. The data is stored in Parquet files, and video files are in MP4 format. The robot type is so101_follower_dragontactile, and the total dataset size is approximately 300MB (100MB for data files and 200MB for video files). It is suitable for training and evaluating robot control models, particularly those involving visual and tactile feedback.




