AttributeError: module 'diffusers' has no attribute 'UNetUnconditionalModel'
#2
by
j35t3r
- opened
I am getting this error: AttributeError: module 'diffusers' has no attribute 'UNetUnconditionalModel' when starting the pipeline...
My package list:
diffusers 0.15.0
denoising-diffusion-pytorch 1.1.1
ema-pytorch 0.1.4
pytorch-fid 0.3.0
pytorch-lightning 1.8.0
torch 2.0.1
torch-tb-profiler 0.4.1
torchmetrics 0.11.4
torchvision 0.15.2
Extended error:
/home/H/lib/python3.9/site-packages/diffusers/pip │
│ elines/pipeline_utils.py:1039 in from_pretrained │
│ │
│ 1036 │ │ │ │ loaded_sub_model = passed_class_obj[name] │
│ 1037 │ │ │ else: │
│ 1038 │ │ │ │ # load sub model │
│ ❱ 1039 │ │ │ │ loaded_sub_model = load_sub_model( │
│ 1040 │ │ │ │ │ library_name=library_name, │
│ 1041 │ │ │ │ │ class_name=class_name, │
│ 1042 │ │ │ │ │ importable_classes=importable_classes, │
│ │
│ /home/H/lib/python3.9/site-packages/diffusers/pip │
│ elines/pipeline_utils.py:367 in load_sub_model │
│ │
│ 364 ): │
│ 365 │ """Helper method to load the module `name` from `library_name` and `class_name`""" │
│ 366 │ # retrieve class candidates │
│ ❱ 367 │ class_obj, class_candidates = get_class_obj_and_candidates( │
│ 368 │ │ library_name, class_name, importable_classes, pipelines, is_pipeline_module │
│ 369 │ ) │
│ 370 │
│ │
│ /home/H/lib/python3.9/site-packages/diffusers/pip │
│ elines/pipeline_utils.py:320 in get_class_obj_and_candidates │
│ │
│ 317 │ │ # else we just import it from the library. │
│ 318 │ │ library = importlib.import_module(library_name) │
│ 319 │ │ │
│ ❱ 320 │ │ class_obj = getattr(library, class_name) │
│ 321 │ │ class_candidates = {c: getattr(library, c, None) for c in importable_classes.key │
│ 322 │ │
│ 323 │ return class_obj, class_candidates