shin1107/eval_koch_default_resnet18_pretrained

该数据集是用于机器人学领域的，包含8个完整的剧集，共有4396帧，1个任务，16个视频和1个数据块，每个数据块的大小为1000。数据集的帧率为30fps，并且训练集的范围是从第0集到第8集。数据以Parquet格式存储，并且包含了动作、状态、顶部和前部图像、时间戳、帧索引、剧集索引、索引和任务索引等特征。

This dataset is for robotics tasks, containing 8 episodes, a total of 4396 frames, 1 task, 16 videos, and 1 chunk with a size of 1000. The dataset has a frame rate of 30fps, and the training set ranges from episode 0 to episode 8. The data is stored in Parquet format and includes features such as actions, states, top and front images, timestamps, frame indices, episode indices, indices, and task indices.