aractingi/tokyo_u_lsmo_dataset

该数据集包含多个特征，如视频帧图像、语言指令、观察状态、动作、时间戳、剧集索引、帧索引、下一个奖励、下一个完成状态和索引。数据集分为训练集，包含955个样本，总大小为218815字节。下载大小为84137字节。

This dataset is primarily used for machine learning tasks, particularly in the field of reinforcement learning. It includes various types of data features such as video frames (for visual observations), language instructions (for task guidance), state sequences (describing system states), action sequences (describing executed actions), timestamps (recording time information), episode indices (distinguishing different episodes), frame indices (distinguishing different frames), next reward (predicted reward value), next done (indicating task completion), and index (unique identifier for data records). The dataset is divided into a training set, containing 955 samples with a total size of 218815 bytes. The dataset configuration is named default, and the data file path is data/train-*.