TerryMelody

Loss is not going down

4

 Hello dear authors! I trained ddim in another dataset. In 1200 epoches,, the loss still seems not going down constantly while sometimes loss became large. I wonder whether it...

high CPU usage discovered in training stage

1

 Hello dear author! Thank you for your contribution. When I trained model on the leather object, I checked 'htop' command on Linux and found that CPU% became very high....

global image representations

4

Hello dear authors！In the code "image_features, patch_tokens = model.encode_image(image, features_list)", is "image_features" the global image representations? Just like the source code in openclip: "image_features = model.encode_image(image)", as we always do....

Hi dear author, thank you for your work! Could you please provide the processed medical-related datasets you mentioned in the paper?

Hello dear authors I have a little question about the model choice and parameter optimization. I set "Human->Zombie", here are the results: a. ViT-B/16: just one epoch can achieve the...

latent diffusion deterministic forward pass

1

In ddim.py, is there any implementation of this todo part: deterministic forward pass? if mask is not None: assert x0 is not None img_orig = self.model.q_sample(x0, ts) # TODO: deterministic...