missing a config.json file for use with standard diffusers
getting this:
OSError Traceback (most recent call last)
/usr/local/lib/python3.10/dist-packages/diffusers/configuration_utils.py in load_config(cls, pretrained_model_name_or_path, return_unused_kwargs, return_commit_hash, **kwargs)
395 )
396 except EntryNotFoundError:
--> 397 raise EnvironmentError(
398 f"{pretrained_model_name_or_path} does not appear to have a file named {cls.config_name}."
399 )
OSError: Nacholmo/controlnet-qr-pattern-sdxl does not appear to have a file named config.json.
The issue was caused by the location of the config file and model in the repository. Should be fixed now
Hooray, thanks!
will test shortly then close this issue.
Uhps! Well, you fixed that error but I'm afraid something is misaligned with my use of the image-to-image pipeline. While it works with other SDXL controlnets and with earlier SD14/1.5 controlnets including monster now it throws this error:
ValueError: AutoPipeline can't find a pipeline linked to StableDiffusionXLImg2ImgControlNetPipeline for None
Thanks again for all your efforts. I hope you can find time and cause to fix this as the SDXL model is far superior. Regardless keep up the good work!
woops, that was the wrong image, should have been this one:
okay! the problem was on my end :) When I switched to the original StableDiffusionXLImg2ImgPipeline instead of the AutoPipeline it ran just fine.