Emm9625/lgbtqi-finepersonas-wip

该数据集包含用户信息和个人喜好的标签，其中用户信息用persona表示，标签用labels表示。matched_categories字段可能表示与用户匹配的分类，has_matches字段则用于指示是否存在匹配的分类。训练集包含大约45万个样本，数据集总大小约为123MB。

The dataset includes user information and personal preference labels, with user information represented by the persona field and labels by the labels field. The matched_categories field might represent the categories that match with the user, while the has_matches field indicates whether there are any matching categories. The training set contains approximately 450,000 samples, with the total dataset size being about 123MB.