Why might a discriminator (for a modified WGAN architecture) with a non-singlular output perform better than traditional discriminator?
the following are the Generator and discriminator model architectures for a (modified) WGAN model that I have come up with for downsampling rainfall data by a factor of 4 (only as a test). The input shape of the dataset is (8030, 14, 21) and the output shape I want is (8030, 28, 42).