김진원

what do you think of returning output of processors as dictionary @ChristiaensBert ```python class MultiProcessor: def __init__(self, processor_id: Union[List[str], str]): # if processor_id is a string, load one processor #...

> @jinwonkim93 , Will you update this pull request? Otherwise, I can open a new pull request. @kadirnar i am not the one who PR this. @vlordier is.

> I think the total_num_steps accounts for the gradient accumulation steps (GAS) somewhere non-obvious (I can't track it down atm). I tried a test training with GAS=1 and it had...

me too but mine is green.

@killnice66 Were you able to converge syncnet to 0.25 under?

you will be able to use it after this PR https://github.com/patrickvonplaten/controlnet_aux/pull/29

Have you check your image? @depyronick

> > Have you check your image? @depyronick > > what do you mean by that? image = base64_to_pil('...base64stringhere') can you check if this is valid image?

> > > > Have you check your image? @depyronick > > > > > > > > > what do you mean by that? > > > > >...

> https://huggingface.co/lllyasviel/ControlNet ```python from diffusers import StableDiffusionControlNetPipeline, ControlNetModel, UniPCMultistepScheduler from controlnet_aux import OpenposeDetector from controlnet_aux.open_pose import Body import torch def base64_to_pil(base64_image_data: str): return Image.open(BytesIO(base64.b64decode(base64_image_data))) image = base64_to_pil('...base64stringhere') pipeline = StableDiffusionControlNetPipeline.from_pretrained(...