Regarding model/ip_sd15_64.bin

#3
by Ling277 - opened

Hi @Adapter I wanna know whether this model is fine-tuned or pretrained. If I understand paper DiffEditor correctly, according to Figure 4 in the paper, Linear, cross attention and feed forward layers are learnable. Can you verify whether my understanding is correct? I look forward to your reply. Thanks

Sign up or log in to comment