Thank you very much for your paper and code!

Thank you very much for your paper and code! However, when I was running it, an error occurred: 

pipeline = MarigoldPipelineRGBLatentNoise.from_pretrained( args.pretrained_model_name_or_path,
text_encoder=text_encoder,
vae=vae,
unet=unet,
revision=args.revision,
variant=args.variant,
), 
it seems that the parameters are insufficient. Could you please tell me how to select the model?