weathon/aas_benchmark-dance_flux

该数据集包含原始和扭曲的图像对，以及对应的提示文本。每个样本都包括图像的原始和扭曲版本、索引、原始和扭曲的提示文本、选择的维度、模型性能指标等特征。数据集适用于训练机器学习模型，特别是那些需要处理图像和文本的模型。

The dataset consists of pairs of original and distorted images along with corresponding prompt texts. Each sample includes features such as the original and distorted versions of the images, an index, the original and distorted prompt texts, selected dimensions, model performance metrics, and more. The dataset is suitable for training machine learning models, especially those that require processing both images and text.