oxe-auge/fractal20220817_data_train_47500_50000_augmented

fractal20220817_data_train_47500_50000_augmented是一个机器人增强数据集，主要用于机器人学习和跨实体策略学习。数据集包含多个机器人的图像和状态数据，如jaco、kinova3、kuka_iiwa、panda、sawyer、ur5e、xarm7等。数据集中包含2500个episodes，108603帧图像，帧率为3fps。数据集提供了丰富的特征，包括机器人的图像、关节位置、末端执行器位置、自然语言指令等。此外，数据集还提供了详细的配置信息和数据布局。

fractal20220817_data_train_47500_50000_augmented is a robotics augmented dataset primarily used for robotics learning and cross-embodiment policy learning. The dataset includes image and state data for multiple robots such as jaco, kinova3, kuka_iiwa, panda, sawyer, ur5e, xarm7, etc. The dataset contains 2,500 episodes, 108,603 frames, with a frame rate of 3fps. It provides rich features including robot images, joint positions, end-effector positions, natural language instructions, etc. Additionally, the dataset offers detailed configuration information and data layout.