Default Width, Height in ControlNet does not initialized

[rootonchair]

Describe the bug

I notice that width, height variables are not initialized as self.unet.config.sample_size * self.vae_scale_factor said in the doc
https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/controlnet/pipeline_controlnet.py#L1024

I'm not sure if this is intentional, else it would cause division error between two None varables when #6360 is merged

Reproduction

None

Logs

No response

System Info

• diffusers version: 0.25.0.dev0
• Platform: Linux-4.18.0-408.el8.x86_64-x86_64-with-glibc2.28
• Python version: 3.9.18
• PyTorch version (GPU?): 2.1.2+cu121 (True)
• Huggingface_hub version: 0.20.1
• Transformers version: 4.36.2
• Accelerate version: 0.25.0
• xFormers version: not installed

Who can help?

@yiyixuxu

[DN6]

cc: @yiyixuxu Would you be able to take a look here please before #6360 is merged

[yiyixuxu]

hi @rootonchair

#6360 does not touch the controlnet pipeline - I assume you are refering to controlnet inpaint pipeline?

it sets width, height variable here

diffusers/src/diffusers/pipelines/controlnet/pipeline_controlnet_inpaint.py

Line 1350 in dd4459a

_, _, height, width = init_image.shape

[rootonchair]

@yiyixuxu yes, I'm refering to controlnet inpaint pipeline. Sorry, I'm a little confused, why don't we initialize it as self.unet.config.sample_size * self.vae_scale_factor?

[yiyixuxu]

@rootonchair good question! sorry I think the docstring might be wrong.
it is not default to self.unet.config.sample_size * self.vae_scale_factor, see the code here:
https://github.com/huggingface/diffusers/blob/3e8b63216e502a0dd7424hec0fef4134187802b93f/src/diffusers/image_processor.py#L401

[rootonchair]

https://github.com/huggingface/diffusers/blob/3e8b63216e502a0dd7424hec0fef4134187802b93f/src/diffusers/image_processor.py#L401

weird, the path doesn't exists

[yiyixuxu]

This?

https://github.com/huggingface/diffusers/blob/main/src/diffusers/image_processor.py#L401

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.