mbodiai/ABB_PandG_20250402_141056

该数据集包含了多个特征，如图像、指令、提示、行动、状态、奖励和元数据等信息。行动中具体记录了姿态、抓握和夹爪力度，状态中包含了图像、深度、世界物体和夹爪姿态。所有数据被分为训练集，共8个样本。数据集总大小为84804730.0字节。

The dataset includes multiple features such as images, instructions, prompts, actions, states, rewards, and metadata. Actions specifically record posture, grasp, and gripper force, while states include images, depths, world objects, and gripper poses. All data is divided into a training set with a total of 8 samples. The dataset is 84804730.0 bytes in size.