BRAZOWSKI/Belgium_Houses_2026

该数据集用于训练VL（视觉语言）模型，数据来源于比利时房屋销售网站immoweb.be。数据集包含房屋广告的标题（法语）、房屋图片、简短描述（法语）、居住面积（平方米）、建筑类型、PEB（能源性能证书）、PEB数值（kWh/m2）、总PEB（kWh/年）、立面数量（2、3或4）、是否有双层玻璃（法语）、环境（郊区、乡村、城市等）、建筑状态、价格（欧元）和广告URL。

Dataset to train VL. This dataset is from a house seller web site (immoweb.be). Its contains The title of the announce (in french), the image of the house, a brief description (in french), the living area in m2, the type of building, the PEB, the PEB number in kWh/m2, the total PEB in kWh/year, the number of facades (2, 3 or 4), if therere double glazing or not (in french), the environnement (suburb, countryside, urban, ...), state of the building, price in euros, and the URL of the announce.