oxe-auge/fractal20220817_data_train_10000_12500_augmented

fractal20220817_data_train_10000_12500_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。数据集包含多种机器人的图像和状态信息，如jaco、kinova3、kuka_iiwa、panda、sawyer、ur5e、xarm7等。数据集的帧率为3fps，总共有2500个episodes和107885帧。数据以parquet格式存储，包含多种特征如机器人图像、关节位置、末端执行器位置等。数据集还提供了自然语言指令，用于描述任务。

fractal20220817_data_train_10000_12500_augmented is a robotics augmentation dataset primarily used for robot learning and cross-embodiment policy learning. The dataset includes images and state information for various robots such as jaco, kinova3, kuka_iiwa, panda, sawyer, ur5e, and xarm7. The dataset has a frame rate of 3fps, with a total of 2,500 episodes and 107,885 frames. The data is stored in parquet format and includes various features such as robot images, joint positions, end-effector positions, etc. The dataset also provides natural language instructions to describe tasks.