leokswang/YAM_lerobot_format

该数据集是一个机器人学习数据集，使用LeRobot框架创建，专门针对yam机器人类型。它包含50个训练episodes，总计101,695帧数据，帧率为30fps。数据以parquet文件格式存储，分为一个chunk。数据集提供了多模态观测数据，包括来自三个摄像头（顶部、左侧、右侧）的图像，每个图像分辨率为360x640像素、3个通道；以及14维的机器人状态观测（浮点类型）。此外，还包含14维的动作数据（浮点类型），以及时间戳、帧索引、episode索引、任务索引等元数据。该数据集适用于机器人控制、强化学习等任务，但未提供主页、论文或引用信息。

This dataset is a robotics learning dataset created using the LeRobot framework, specifically for the yam robot type. It contains 50 training episodes, totaling 101,695 frames at a frame rate of 30fps. The data is stored in parquet format and organized into one chunk. The dataset provides multimodal observation data, including images from three cameras (top, left, right), each with a resolution of 360x640 pixels and 3 channels, as well as 14-dimensional robot state observations (float32 type). Additionally, it includes 14-dimensional action data (float32 type), along with metadata such as timestamps, frame indices, episode indices, and task indices. The dataset is suitable for tasks like robot control and reinforcement learning, but no homepage, paper, or citation information is provided.