gwang-kim / DiffusionCLIP

[CVPR 2022] Official PyTorch Implementation for DiffusionCLIP: Text-guided Image Manipulation Using Diffusion Models
Other
785 stars 113 forks source link

Different results in ViT-B/16 and ViT-L/14@336px #36

Open TerryMelody opened 5 months ago

TerryMelody commented 5 months ago

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 good result like follows image b. ViT-L/14@336px: but when I use this one the results seem strange. I don't know whether it should be set different parameters to finetune the diffusion model. (left->right:1-5 epochs) image image image image image