eval_so101_grab_block_act_r0

该数据集是使用LeRobot创建的机器人操作数据集，专门针对so_follower机器人类型。数据集包含5个episodes，总计1789帧数据，帧率为30fps。数据以parquet格式存储，总数据文件大小为100MB，视频文件大小为200MB。数据集特征包括动作（action）、观测状态（observation.state）、前视图像视频（observation.images.front）以及补充信息如策略动作、干预状态、收集器策略ID等。动作和状态特征涉及6个关节位置：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。视频数据分辨率为480x640，3通道，使用av1编解码器。数据集仅包含训练分割，用于机器人学习和控制任务。This dataset is a robot manipulation dataset created using LeRobot, specifically for the so_follower robot type. It contains 5 episodes, totaling 1789 frames of data at a frame rate of 30fps. The data is stored in parquet format, with a total data file size of 100MB and a video file size of 200MB. Dataset features include actions, observation states (such as joint positions), front-view image videos, and supplementary information like policy actions, intervention status, and collector policy IDs. The action and state features involve 6 joint positions: shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position. The video data has a resolution of 480x640 with 3 channels and uses the av1 codec. The dataset only includes a training split and is intended for robot learning and control tasks.