mbodiai/ABB_PandG_20250402_153228

该数据集包含了多个特征，如图像、指令、提示、位姿、抓取力度、夹持器力量等。它还包括状态信息，如前后摄像头图像、深度信息、世界中的物体和夹持器姿态。每个样本还有一个浮点数表示的奖励和元数据，其中包含剧集索引和步索引。数据集分为训练集，共有6个样本，总大小约为63.7MB。

The dataset includes multiple features such as images, instructions, prompts, poses, grasping force, and gripper force. It also contains state information such as front and rear camera images, depth information, objects in the world, and gripper pose. Each sample has a reward represented by a floating-point number and metadata including episode index and step index. The dataset is split into a training set with a total of 6 samples, with a size of approximately 63.7MB.