TEST SAMPLE CATEGORY DETERMINATION METHOD AND APPARATUS FOR SIAMESE NETWORK, AND TERMINAL DEVICE

摘要

Provided are a test sample category determination method and apparatus for a Siamese network, and a terminal device, which are applicable to the field of digital medical treatment. The method comprises: acquiring a vector representation of any sample of any category in training samples by means of a Siamese network (S200); acquiring the weight of said any sample in said any category (S201); determining a category representation of said any category according to the vector representation of each sample in said any category and the weight of each sample (S202); acquiring a sample vector representation of a test sample by means of the Siamese network (S203); constructing sample pairs according to the sample vector representation and the category representations of all the categories in the training samples (S204); inputting each sample pair into the Siamese network, and determining the similarity between each category representation and the sample vector representation by means of the Siamese network (S205); and determining the category of the test sample on the basis of the similarity between each category representation and the sample vector representation (S206). By using the present method, the classification precision of the category of a test sample in a Siamese network can be improved.