'BitDiffusion' object has no attribute 'scheduler'

[BioGeek]

Describe the bug

Running the Bit Diffusion example code from the Community Pipeline Examples fails

Reproduction

from diffusers import DiffusionPipeline
pipe = DiffusionPipeline.from_pretrained("google/ddpm-cifar10-32", custom_pipeline="bit_diffusion")
image = pipe().images[0]

Logs

AttributeError Traceback (most recent call last)
<ipython-input-2-3d32a9d66392> in <cell line: 2>()
 1 from diffusers import DiffusionPipeline
----> 2 pipe = DiffusionPipeline.from_pretrained("google/ddpm-cifar10-32", custom_pipeline="bit_diffusion")
 3 image = pipe().images[0]

/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_validators.py in _inner_fn(*args, **kwargs)
 112 kwargs = smoothly_deprecate_use_auth_token(fn_name=fn.__name__, has_token=has_token, kwargs=kwargs)
 113 
--> 114 return fn(*args, **kwargs)
 115 
 116 return _inner_fn # type: ignore

/usr/local/lib/python3.10/dist-packages/diffusers/pipelines/pipeline_utils.py in from_pretrained(cls, pretrained_model_name_or_path, **kwargs)
 964 
 965 # 10. Instantiate the pipeline
--> 966 model = pipeline_class(**init_kwargs)
 967 
 968 # 11. Save where the model was instantiated from

~/.cache/huggingface/modules/diffusers_modules/git/bit_diffusion.py in __init__(self, unet, scheduler, bit_scale)
 220 super().__init__()
 221 self.bit_scale = bit_scale
--> 222 self.scheduler.step = (
 223 ddim_bit_scheduler_step if isinstance(scheduler, DDIMScheduler) else ddpm_bit_scheduler_step
 224 )

/usr/local/lib/python3.10/dist-packages/diffusers/configuration_utils.py in __getattr__(self, name)
 141 return self._internal_dict[name]
 142 
--> 143 raise AttributeError(f"'{type(self).__name__}' object has no attribute '{name}'")
 144 
 145 def save_config(self, save_directory: Union[str, os.PathLike], push_to_hub: bool = False, **kwargs):

AttributeError: 'BitDiffusion' object has no attribute 'scheduler'

System Info

• 🤗 Diffusers version: 0.30.0
• Platform: Linux-6.1.85+-x86_64-with-glibc2.35
• Running on Google Colab?: Yes
• Python version: 3.10.12
• PyTorch version (GPU?): 2.3.1+cu121 (False)
• Flax version (CPU?/GPU?/TPU?): 0.8.4 (cpu)
• Jax version: 0.4.26
• JaxLib version: 0.4.26
• Huggingface_hub version: 0.23.5
• Transformers version: 4.42.4
• Accelerate version: 0.32.1
• PEFT version: not installed
• Bitsandbytes version: not installed
• Safetensors version: 0.4.3
• xFormers version: not installed
• Accelerator: NA
• Using GPU in script?: no
• Using distributed or parallel set-up in script?: no

Who can help?

@patrickvonplaten (78a6eed) @kingstut (#971)

[a-r-r-o-w]

I believe this pipeline may have been incorrectly written and isn't well tested because community pipelines are maintained by the community members and don't necessarily see any high priority involvement from us.

I think the fix here would be to move the call to self.register_modules above self.scheduler.step = ..., but not sure and don't have the bandwidth to test and fix. If this helps or if you know of a way fix this, feel free to PR :)

diffusers/examples/community/bit_diffusion.py

Line 226 in 8e3affc

self.register_modules(unet=unet, scheduler=scheduler)

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.