Wasserstein-AutoEncoders

Wasserstein-AutoEncoders copied to clipboard

I think MMD approximation with batch size n should be dists_x divided by n(n-1) (analogous for dists_y) and dists_c divided by n**2. But this Implementation only divides with n-1 and...

DaehanKim

backward MMD

1

Hello Schelotto I have a short question ... why total_loss = recon_loss - mmd_loss total_loss.backward() and not + mmd_loss as both should be descended in the optimization step ? Little...

adrienchaton

I can’t download the source code, or the file is damaged when the download is complete. I don't know if there is something wrong with the source file, or is...

FOURIERWANG

Hi, Thanks for your shared codes and I learned so much out of it! I'm going to modify your wae implementation for 1D sequential data (protein sequences reconstruction), However, your...

payman-ap