ziyu3141/rf_newtrain_2_12

该数据集包含了图像文件的美学评估信息，包括美学分数、人工制品分数、对齐错误分数和总分等。同时，提供了人工制品和对齐错误的热图，以及对齐错误的标记信息。还包括了图像是否均匀、是否被选为首选等布尔类型字段，以及偏好图像、非偏好图像和修订图像的二进制表示。数据集分为训练集，可用于图像美学评估相关的机器学习任务。

This dataset contains aesthetic evaluation information of image files, including aesthetics score, artifact score, misalignment score, and overall score. It also provides heatmaps for artifacts and misalignments, as well as label information for misalignment tokens. Boolean fields include whether the image is uneven, whether it is preferred, and binary representations of preferred image, unpreferred image, and revised image. The dataset is split into a training set, which can be used for machine learning tasks related to image aesthetic evaluation.