Hi, I am feeling stuck at this point because I can not to find which is the issue. The following is my code in the get_gen_loss function:
{moderator edit - solution code removed}
It is simple, but I got an “error” in the assert:
—> 21 assert torch.all(torch.abs(gen_loss_tensor - 1) < 1e-5)
Could anyone help me, please? Which is the problem in the code?
Hi @carlosmera20 ,
Welcome to the community! This is your first post 
Regarding your error, please read again the rule on the loss calculation:
“Remember the generator wants the discriminator to think that its fake images are real”
Right now you are passing zeros_like… may be try passing an argument that makes the discriminator think that the fakes are real.
Please try that and let me know how it goes.
Juan