How is referencenet(latents_ref_img, ref_timesteps, encoder_hidden_states) in Line 492 in train.py connected to the loss?

guoqincode / Open-AnimateAnyone

Unofficial Implementation of Animate Anyone

2.91k stars 233 forks source link

Open souvikg544 opened 6 months ago

souvikg544 commented 6 months ago

There is no variable thats accepting the response of referencenet neither am I sending the sample to any other network. How Does it work?

souvikg544 commented 6 months ago

hello ... Can the dev team explain this one?