Bybychen

Thanks for your reply. I see the point of minimizing the d_loss, but I still get confused of maximizing EM-distance and minimizing EM-distance. During training we minimize d_loss to obtain...

> But in theory the EM-distance will eventually go down, with both effects from updating critic and generator. Like in the raw GAN ? This sort of curves are also...

This helps a lot. Thanks!

Thank a lot!

I have tried a lot of times to visualize esim in docker, but unfortunately, it seems to be infeasible. Maybe you should build esim from source instead of using docker....