rahulunair / stable_diffusion_arc Goto Github PK

I think there is missing line in _optimize_pipeline, the function does not return anything, so the code

    def optimize_pipeline(self) -> None:
        """
        Optimizes the current model pipeline.
        """

        self.pipeline = self._optimize_pipeline(self.pipeline)

is actually setting self.pipeline to None.

(sd_arc) xxx@ubuntu:~/sd_arc/stable_diffusion_arc/text_to_image$ python sd_text_to_img.py
Available models are:

1. runwayml/stable-diffusion-v1-5
2. CompVis/stable-diffusion-v1-4
3. stabilityai/stable-diffusion-2-1
4. stabilityai/stable-diffusion-xl-base-1.0
   Select a model by entering its number (or press enter to use the default model): 1
   Loading the model...
   Loading pipeline components...: 71%|█████████████████████████████████████████████████████████████████████████████████████████████████▏ | 5/7 [00:00<00:00, 9.93it/s]text_config_dict is provided which will be used to initialize CLIPTextConfig. The value text_config["id2label"] will be overriden.
   text_config_dict is provided which will be used to initialize CLIPTextConfig. The value text_config["bos_token_id"] will be overriden.
   text_config_dict is provided which will be used to initialize CLIPTextConfig. The value text_config["eos_token_id"] will be overriden.
   Loading pipeline components...: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:00<00:00, 8.16it/s]
   Model loaded.
   Optimizing the model...
   Traceback (most recent call last):
   File "/home/intel/sd_arc/stable_diffusion_arc/text_to_image/sd_text_to_img.py", line 216, in
   model = Text2ImgModel(model_id, device="xpu")
   File "/home/intel/sd_arc/stable_diffusion_arc/text_to_image/sd_text_to_img.py", line 59, in init
   self.optimize_pipeline()
   File "/home/intel/sd_arc/stable_diffusion_arc/text_to_image/sd_text_to_img.py", line 142, in optimize_pipeline
   self.pipeline = self._optimize_pipeline(self.pipeline)
   File "/home/intel/sd_arc/stable_diffusion_arc/text_to_image/sd_text_to_img.py", line 108, in _optimize_pipeline
   if isinstance(getattr(pipeline, attr), nn.Module):
   File "/home/intel/anaconda3/envs/sd_arc/lib/python3.10/site-packages/diffusers/configuration_utils.py", line 138, in getattr
   raise AttributeError(f"'{type(self).name}' object has no attribute '{name}'")
   AttributeError: 'StableDiffusionPipeline' object has no attribute 'clip_skip'