something wrong when changing CLIP
aigias opened this issue · 2 comments
Hi there,
I found something seems to be wrong when I change CLIP of a model.
My process as below:
- load model1, advance bar - class=CLIP-v1 then "Export". Got model1.clip.pt in components folder.
- load model2, advanced bar - class=CLIP-v1, then choose model1.clip.pt from the "file" drop down menu, then "import".
The webui page started to work and stopped in a few seconds. No file is generated and I got nothing in the webui console.
Here is the report from the webui:
Report (1794/1614/0000)
Statistics
Total keys: 1145 (1.99 GB).
Architecture
SD-v1
UNET-v1
UNET-v1-SD
CLIP-v1
CLIP-v1-SD
VAE-v1
VAE-v1-SD
Rejected
UNET-v1-Inpainting: Missing required keys (1 of 686)
model.diffusion_model.input_blocks.0.0.weight (320, 9, 3, 3)
UNET-v1-Pix2Pix: Missing required keys (1 of 686)
model.diffusion_model.input_blocks.0.0.weight (320, 8, 3, 3)
UNET-v2-SD: Missing required keys (64 of 686)
model.diffusion_model.middle_block.1.transformer_blocks.0.attn2.to_k.weight (1280, 1024)
model.diffusion_model.output_blocks.9.1.proj_out.weight (320, 320)
model.diffusion_model.input_blocks.8.1.proj_out.weight (1280, 1280)
model.diffusion_model.input_blocks.5.1.transformer_blocks.0.attn2.to_k.weight (640, 1024)
model.diffusion_model.input_blocks.4.1.proj_out.weight (640, 640)
…
UNET-v2-Inpainting: Missing required keys (65 of 686)
model.diffusion_model.middle_block.1.transformer_blocks.0.attn2.to_k.weight (1280, 1024)
model.diffusion_model.output_blocks.9.1.proj_out.weight (320, 320)
model.diffusion_model.input_blocks.8.1.proj_out.weight (1280, 1280)
model.diffusion_model.input_blocks.5.1.transformer_blocks.0.attn2.to_k.weight (640, 1024)
model.diffusion_model.input_blocks.4.1.proj_out.weight (640, 640)
…
UNET-v2-Depth: Missing required keys (65 of 686)
model.diffusion_model.middle_block.1.transformer_blocks.0.attn2.to_k.weight (1280, 1024)
model.diffusion_model.output_blocks.9.1.proj_out.weight (320, 320)
model.diffusion_model.input_blocks.8.1.proj_out.weight (1280, 1280)
model.diffusion_model.input_blocks.5.1.transformer_blocks.0.attn2.to_k.weight (640, 1024)
model.diffusion_model.input_blocks.4.1.proj_out.weight (640, 640)
…
SD-v1-Pix2Pix: Missing required classes
UNET-v1-Pix2Pix
SD-v1-ControlNet: Missing required classes
ControlNet-v1
SD-v2: Missing required classes
UNET-v2
CLIP-v2
SD-v2-Depth: Missing required classes
UNET-v2-Depth
Depth-v2
CLIP-v2
By the way, i used the same process but changing VAE. It worked as expected. New file is generated in stable-duffusion folder. Not sure it's a bug or simply it cannot be changed like that. I'm quite new to AIGC basic theory.
Thanks!
I just realized if it's because the import has successed and I just need to click SAVE to generate the new file...
ok my bad. save file and done. sorry with this mess.