Disjoint 10 1

Disjoint 10-1 is an evaluation method in the field of computer vision, designed to rigorously test and assess a model's generalization ability on unseen data by strictly dividing categories. This method splits the dataset into two completely disjoint parts, one for training and the other for testing, ensuring that the model's performance on new categories is more genuinely reliable. It has significant application value in enhancing the robustness and adaptability of models, especially in scenarios such as open-world recognition and zero-shot learning.