tingtingou/rf_train_100_inst_new

该数据集包含多个特征，如文件名、美学评分、伪影评分、错位评分、总体评分、伪影热图、错位热图、错位标记、提示、首选图像、非首选图像和指令等。数据集分为训练集，包含90个样本，总大小为595,893,368字节。下载大小为30,888,875字节。

This dataset is primarily used for evaluating and analyzing the aesthetic quality, artifacts, and alignment of images. It includes features such as Filename, Aesthetics score, Artifact score, Misalignment score, Overall score, Artifact heatmap, Misalignment heatmap, Misalignment token label, prompt, image_preferred, image_not_preferred, and instruction. The dataset is split into a training set with 90 examples. The size of the dataset and the download size are also clearly specified.