Cooking State Classification Dataset

该数据集包含了15种烹饪对象的图片以及11个状态类别，旨在用于识别与烹饪相关的图片中的状态。此外，该数据集还标注了对象标签，并通过数据增强技术来平衡各个类别。规模上，该数据集大约有9.5千张图片，分为70%的训练集、15%的验证集以及15%的测试集。其任务是针对烹饪图片中的对象和状态进行分类。

This dataset comprises images of 15 types of cooking objects and 11 state categories, designed for state recognition in cooking-related images. Additionally, object labels are annotated for the dataset, and data augmentation techniques are utilized to balance the distribution across all categories. In terms of scale, the dataset contains roughly 9,500 images, which are divided into a 70% training split, 15% validation split, and 15% test split. The core task of this dataset is to classify both the objects and states present in cooking images.