ly_grasp_4obj_50_s3

该数据集是使用LeRobot创建的机器人学数据集，主要用于机器人模仿学习或强化学习任务。数据集包含50个episodes，总计29907帧，涉及4个不同的任务。数据以parquet格式存储，视频文件以mp4格式存储。数据集包括以下特征：动作数据（7维浮点数组，表示机器人关节位置和夹爪位置）、观测状态（7维浮点数组，表示关节位置和夹爪位置）、正面和侧面图像观测（视频格式，分辨率480x640，3通道，30fps）、时间戳、帧索引、episode索引、任务索引等。数据集的分辨率为480x640，帧率为30fps，视频编码为av1，无音频。数据集总大小为100MB（数据文件）和200MB（视频文件）。数据集仅包含训练集，用于机器人控制任务的训练和评估。This robotics dataset was created using LeRobot, primarily intended for robot imitation learning or reinforcement learning tasks. It contains 50 episodes, totaling 29907 frames, covering 4 distinct tasks. The dataset is stored in Parquet format, while the video files are in MP4 format. The dataset includes the following features: action data (7-dimensional floating-point array representing robot joint positions and gripper positions), observation states (7-dimensional floating-point array representing joint positions and gripper positions), front and side image observations (video format with resolution 480x640, 3 channels, 30fps), timestamps, frame indices, episode indices, task indices, etc. The dataset has a resolution of 480x640, a frame rate of 30fps, uses AV1 video encoding, and contains no audio. The total size of the dataset is 100MB for data files and 200MB for video files. This dataset only includes a training set, which is used for the training and evaluation of robot control tasks.