-
Notifications
You must be signed in to change notification settings - Fork 533
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Inpainting #19
Comments
The inpainting controlnet example seems incomplete as it doesn’t actually inpaint. Maybe some code is missing there. |
Is this issue solved? |
I did some experiments, and it seems that inpainting still not works well.. |
Note that this only fixes the missing images. The controlnet example is otherwise still broken. |
Any updates regarding the inpainting example? |
There have been no updates in the last two months unfortunately. The authors are busy with other things. |
at this section in the example notebook:
`batch_size = 4
url = "https://cdn.discordapp.com/attachments/1121232062708457508/1204787053892603914/cat_dog.png?ex=65d60061&is=65c38b61&hm=37c3d179a39b1eca4b8894e3c239930cedcbb965da00ae2209cca45f883f86f4&"
images = resize_image(download_image(url)).unsqueeze(0).expand(batch_size, -1, -1, -1)
batch = {'images': images}
mask = None
mask = torch.ones(batch_size, 1, images.size(2), images.size(3)).bool()
outpaint = False
threshold = 0.2
with torch.no_grad(), torch.cuda.amp.autocast(dtype=torch.bfloat16):
cnet, cnet_input = core.get_cnet(batch, models, extras, mask=mask, outpaint=outpaint, threshold=threshold)
cnet_uncond = cnet
show_images(batch['images'])
show_images(cnet_input)`
i received this error:
`---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
in <cell line: 13>()
12
13 with torch.no_grad(), torch.cuda.amp.autocast(dtype=torch.bfloat16):
---> 14 cnet, cnet_input = core.get_cnet(batch, models, extras, mask=mask, outpaint=outpaint, threshold=threshold)
15 cnet_uncond = cnet
16
/content/StableCascade/train/train_c_controlnet.py in get_cnet(self, batch, models, extras, cnet_input, **kwargs)
141 with torch.no_grad():
142 if cnet_input is None:
--> 143 cnet_input = extras.controlnet_filter(images, **kwargs)
144 if isinstance(cnet_input, tuple):
145 cnet_input, cnet_input_preview = cnet_input
TypeError: SREffnetFilter.call() got an unexpected keyword argument 'mask'`
and am unable to generate the masks for the next steps.
how should i fix this?
thanks for providing the notebooks, my text2image/image2image colab version runs great. I'm trying to build out the remaining features.
The text was updated successfully, but these errors were encountered: