Hello, I was trying to reproduce the paper's experiments for AMD classification. While I am getting AUC scores similar to those reported in the paper, the metric curves for validation and test seem very unstable, and the loss keeps increasing. Did you observe the same issue when running the experiments?
Hello, I was trying to reproduce the paper's experiments for AMD classification. While I am getting AUC scores similar to those reported in the paper, the metric curves for validation and test seem very unstable, and the loss keeps increasing. Did you observe the same issue when running the experiments?