Rbrq03 comments

Results 16 comments of


                                            Rbrq03

Unsuccessful cross-attention weight loading in Custom Diffusion with StableDiffusionAttendAndExcitePipeline

Combing them seems a good idea. Could you share some images generated by this piece of code? I am not sure what it looks like.

Unsuccessful cross-attention weight loading in Custom Diffusion with StableDiffusionAttendAndExcitePipeline

hey @daeunni, maybe #7294 helps

Fix custom_diffusion with PEFT installed issue#7261

Hey Saya, need a review for the new commit, I reformat this file @sayakpaul

Fix custom_diffusion with PEFT installed issue#7261

I am sorry for that.😂 What happens to this commit? Should I do something to help it pass the CI?

Unsuccessful cross-attention weight loading in Custom Diffusion

I open a PR for fix this problem. In my local test, cross-attention weights can be loaded successfully in my local test

Unsuccessful cross-attention weight loading in Custom Diffusion

@daeunni you can try one of these methods: 1. simply uninstall PEFT if you don't use lora. 2. modify code in diffusers/loaders/unets.py as my PR do I hope it will...

Unsuccessful cross-attention weight loading in Custom Diffusion

Glad to hear that :)

Can I use this with colab?

I think the answer is yes, you can use lite.yaml as your config which may use model online. You need not to deploy and download any model.You can copy all...

TypeError: Linear.forward() got an unexpected keyword argument 'scale'

Everyone, presently, this issue can be effectively addressed through the following steps: `pip uninstall diffusers ` `pip install diffusers==0.18.1` It works for me, hope it helps

issues on train.py

It appears that the pretrained weights may not have been loaded correctly. You can try checking the 'pretrained_model_path' key in the config. Additionally, there might be an issue with the...