weidler / RLaSpa

Reinforcement Learning in Latent Space

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Implement SiameseAutoencoder

weidler opened this issue · comments

Ok I have the scheleton of the network but I'm stuck on the loss function. I have to use two different loss or if it's not possible in pycharm I have to combine them. Anyone has a clue on how to do it? I remember @dannigt said something about a previous project in which they were combining two loss, do I remember correctly?