Now the code generates negative samples in advance every time, and then shuffles the order, and then randomly selects from the set of positive and negative samples for training. If I extract certain positive datas for each training and generate the corresponding negative samples, that is, each batch positive sample has the corresponding negative sample, is the training discriminator feasible?
Now the code generates negative samples in advance every time, and then shuffles the order, and then randomly selects from the set of positive and negative samples for training. If I extract certain positive datas for each training and generate the corresponding negative samples, that is, each batch positive sample has the corresponding negative sample, is the training discriminator feasible?