ziyu3141/rf_newtrain_8_49

该数据集包含了图片的相关信息，如审美评分、瑕疵评分、对齐情况评分等，并提供了与之相关的热图和标记。数据集中的图片分为首选、非首选和修订三类。此外，数据集还包含了未修订图片的ID信息。数据集总共包含100个示例，并分为训练集。数据集的总大小为677,402,855字节。

The dataset includes image-related information such as aesthetics scores, artifact scores, misalignment scores, along with associated heatmaps and labels. Images in the dataset are categorized into preferred, unpreferred, and revised types. Additionally, the dataset contains IDs for unrevised images. The dataset consists of 100 examples in total and is split into a training set. The overall size of the dataset is 677,402,855 bytes.