I was building the training pipeline for a siamese network for one-shot learning. Everybody on the internet uses accuracy rather than loss for stopping training. Why use one-shot accuracy over loss value for stopping? Aren't the concepts of overfitting and underfitting relevant here or am I missing something fundamental in one-shot learning?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.