shubhamt0802/all_tasks_merged

该数据集是使用LeRobot创建的，专门用于机器人学任务。它包含157个总片段，117270个总帧数，覆盖7个任务。数据集结构包括动作、观察状态、图像特征（如前视和手腕摄像头视频）以及时间戳、帧索引、片段索引等元数据。动作和观察状态特征均包含6个浮点数值，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）。图像特征以视频形式存储，分辨率为480x640，3通道，帧率为30fps，使用av1编解码器。数据集总大小为300MB（数据文件100MB，视频文件200MB），采用Apache-2.0许可证。机器人类型为so101_follower，数据以parquet文件格式组织，并包含训练分割。This dataset was created using LeRobot and is designed for robotics tasks. It includes a total of 157 episodes, 117270 frames, and covers 7 tasks. The dataset structure comprises actions, observation states, image features (e.g., front and wrist camera videos), and metadata such as timestamps, frame indices, and episode indices. Both action and observation state features contain 6 float32 values corresponding to robot joint positions (e.g., shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper positions). Image features are stored as videos with a resolution of 480x640, 3 channels, a frame rate of 30fps, and use the av1 codec. The total dataset size is 300MB (100MB for data files and 200MB for video files), licensed under Apache-2.0. The robot type is so101_follower, data is organized in parquet files, and includes a training split.