Open Buddies-as-you-know opened 1 year ago
Does MIN2Net do the classification like encoder->decoder->softmax? My understanding of MIN2Net without decoder is that the structure is encoder->softmax.
Yes, but both also have metric learning at the latent vector.
Does MIN2Net do the classification like encoder->decoder->softmax? My understanding of MIN2Net without decoder is that the structure is encoder->softmax.