Nulia/vista-mini

Vista Mini数据集包含图像和文本描述两种数据类型，分为训练集和测试集。训练集包含1916个样本，测试集包含20个样本。该数据集主要用于图像分类和图像到文本的任务，语言为英语，标签包括不适合所有观众、艺术和合成。数据集来源于CivitAI，描述由Joytag创建，主要用于视觉语言模型微调的描述任务。

The dataset contains images and corresponding caption texts. The image data type is image, and the caption text data type is string. The dataset is divided into training set and test set, with 1916 samples in the training set and 20 samples in the test set. The dataset is suitable for image classification and image-to-text tasks. The language of the dataset is English, with tags including not-for-all-audiences, art, synthetic. The name of the dataset is Vista Mini, and the size category is 1K<n<10K. The source of the dataset is CivitAI, and the caption texts are created using Joytag, primarily for VLM fine-tuning.