guloyy/SO101-eval2-brg-expanded-v3-256

该数据集使用LeRobot创建，专注于机器人任务，包含300个episodes、134050帧和165个任务。数据以parquet文件存储，总数据文件大小为100MB，视频文件大小为200MB，帧率为30fps。数据集结构包括动作、观察状态、前视图像、时间戳、帧索引、episode索引、索引和任务索引等特征。动作和观察状态特征均为6维浮点数组，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）。前视图像特征为256x256x3的视频数据，使用av1编解码器。数据集仅包含训练分割，适用于机器人控制和模仿学习任务。

This dataset was created using LeRobot and focuses on robotics tasks. It contains 300 episodes, 134050 frames, and 165 tasks. The data is stored in parquet files, with a total data file size of 100MB and video file size of 200MB, at a frame rate of 30fps. The dataset structure includes features such as action, observation state, front images, timestamp, frame index, episode index, index, and task index. The action and observation state features are both 6-dimensional float arrays corresponding to robot joint positions (e.g., shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper positions). The front image feature is video data of size 256x256x3, using the av1 codec. The dataset only includes a training split and is suitable for robotics control and imitation learning tasks.