TatvaRA/skin-lesion-preference-dataset

该数据集包含两个实体（A和B）的多种特征和嵌入数据，包括mask_features、gradcam_features、mask_text_embeds和gradcam_text_embeds，以及偏好（preference）、奖励A（reward_A）和奖励B（reward_B）字段。数据集分为一个训练集，包含240个样本。

This dataset contains various features and embeddings for two entities (A and B), including mask_features, gradcam_features, mask_text_embeds, and gradcam_text_embeds, along with preference, reward_A, and reward_B fields. The dataset is split into a single training set with 240 examples.