Mysterybullet

icon company qz icon location sh icon location nothing just pure

Results 4 comments of Mysterybullet

How to embed lora in a custom stable diffusion network

mark

question about Vb loss

However, the code include model_kwargs in _vlb_terms_bpd while choose the KL loss, so i guess whether the model_kwargs is forgetten in the MSE loss (learn_sigma).

Question about variance calculation in sampling step

I have the same doubt, what's the reason for doing this ?

Is there a non-stochastic sampling method?

care about this too