Open 66ling66 opened 3 months ago
When I montage them,I just get this image.Do you make the same experiments?
yes,I meet the same problem. I have a question that why are the UNet parameters frozen during training for SD1v5, but not for SDXL? the huggingface training script sets "Unet.train()" ???
When I montage them,I just get this image.Do you make the same experiments?![166](https://github.com/TencentARC/T2I-Adapter/assets/108187964/8d7cdcc7-7af2-4c01-adbe-20da2d0ec957)