You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, when using the workflow at first it works fine but it throws the warning in console: WARNING SHAPE MISMATCH diffusion_model.input_blocks.0.0.weight WEIGHT NOT MERGED torch.Size([320, 8, 3, 3]) != torch.Size([320, 4, 3, 3]) Merged with diffusion_model.input_blocks.0.0.weight channel changed from torch.Size([320, 4, 3, 3]) to [320, 8, 3, 3]
it completes both steps correctly (relighting and then outpainting)
when using it the second time it throws the error: WARNING SHAPE MISMATCH diffusion_model.input_blocks.0.0.weight WEIGHT NOT MERGED torch.Size([320, 8, 3, 3]) != torch.Size([320, 4, 3, 3]) 0%| | 0/25 [00:00<?, ?it/s] !!! Exception during processing!!! Given groups=1, weight of size [320, 4, 3, 3], expected input[2, 8, 96, 64] to have 4 channels, but got 8 channels instead Traceback (most recent call last):
stops at the KSampler of the relight workflow.
Steps to reproduce:
Load the workflow into comfy UI
Start the workflow and observe first warning in console
Wait until the workflow ends
Start the workflow second time and it throws the error, needs to restart whole comfy UI to be able to run the workflow again.
Tested evnrionments:
Python 3.10 - comfy stadalone
Python 3.11.8 - comfy portable
Python 3.11.9 - comfy stadalone
in each enviornment I've made sure to install necessary nodes and recomended nodes (ComfyUI-layerdiffuse, ComfyUI-KJNodes,ComfyUI-Easy-Use, ComfyUI_essentials,ComfyUI-IC-Light)
Tried some simple debugging but unable to find anything for now.
Hi, when using the workflow at first it works fine but it throws the warning in console:
WARNING SHAPE MISMATCH diffusion_model.input_blocks.0.0.weight WEIGHT NOT MERGED torch.Size([320, 8, 3, 3]) != torch.Size([320, 4, 3, 3]) Merged with diffusion_model.input_blocks.0.0.weight channel changed from torch.Size([320, 4, 3, 3]) to [320, 8, 3, 3]
it completes both steps correctly (relighting and then outpainting)
when using it the second time it throws the error:
WARNING SHAPE MISMATCH diffusion_model.input_blocks.0.0.weight WEIGHT NOT MERGED torch.Size([320, 8, 3, 3]) != torch.Size([320, 4, 3, 3]) 0%| | 0/25 [00:00<?, ?it/s] !!! Exception during processing!!! Given groups=1, weight of size [320, 4, 3, 3], expected input[2, 8, 96, 64] to have 4 channels, but got 8 channels instead Traceback (most recent call last):
stops at the KSampler of the relight workflow.
Steps to reproduce:
Tested evnrionments:
in each enviornment I've made sure to install necessary nodes and recomended nodes (ComfyUI-layerdiffuse, ComfyUI-KJNodes,ComfyUI-Easy-Use, ComfyUI_essentials,ComfyUI-IC-Light)
Tried some simple debugging but unable to find anything for now.
I've tested the same outpainting method but instead of relighting it with this repository nodes I've used this workflow and combined it with the outpainting workflow, it didint throw any errors or warnings in the console.
https://github.com/kijai/ComfyUI-IC-Light-Wrapper/blob/main/examples/IC-Light_example_spotlight_01.json
this one works fine also
https://github.com/kijai/ComfyUI-IC-Light
(JSON and PNG workflow included below)
workflow-relight-outpaint.json
working-relight-outpainting.json
The text was updated successfully, but these errors were encountered: