Yanze Wu
Yanze Wu
I see many code like this:`fake_data = G(noise).detach()` But in your most code, there is no detach, so is it just ok to omit it ?
After run about 15 epoch, the generator seems to collapse to a parameter setting where it always emits the same point. Did anyone run the script successfully?
I think it is weird, can you explain that?
The Upblock layer index should be [0, 36] rather than [0, 30]?