Vikhrmodels/LLaVa_textcaps_rus

Textcaps数据集包含英文和俄文的图像描述。俄文描述是从英文翻译过来的。原始数据来源于task textcaps，数据集格式包括图像的字节数据、包含图像描述的简短对话（英文和俄文），以及数据来源标签。数据集适用于图像到文本的任务。

The Textcaps dataset includes image descriptions in both English and Russian. The Russian descriptions are translations from the English ones. The original data comes from the task textcaps, and the dataset format includes the byte data of the image, short dialogues with image descriptions (in both English and Russian), and a data source label. The dataset is suitable for image-to-text tasks.