franka_pnp_language

该数据集是使用LeRobot创建的，专为机器人任务设计。数据集基于LeRobot v2.1代码库，涉及franka机器人类型，包含20个episodes、7673帧和2个任务，帧率为20fps。数据以parquet格式存储，特征包括图像观测（如256x256x3的图像和手腕图像）、状态观测（7维浮点数组）、动作（8维浮点数组）、对象初始化状态（9维浮点数组）以及时间戳、帧索引、episode索引、索引和任务索引等元数据。数据集结构支持训练分割，用于机器人学习和控制任务。

This dataset is created using LeRobot and is specifically designed for robot tasks. It is based on the LeRobot v2.1 codebase, involves the franka robot type, contains 20 episodes, 7673 frames, and 2 tasks, with a frame rate of 20fps. The data is stored in parquet format, with features including image observations (such as 256x256x3 images and wrist images), state observations (7-dimensional floating-point arrays), actions (8-dimensional floating-point arrays), object initialization states (9-dimensional floating-point arrays), and metadata such as timestamps, frame indices, episode indices, indices, and task indices. The dataset structure supports training splits for robot learning and control tasks.