TypeError when loading pipeline

Hello everyone,

I’m trying to follow these instructions : Loading pipelines that require access request

But I get this :

from diffusers import DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained(“runwayml/stable-diffusion-v1-5”)
Fetching 19 files: 100%|███████████████████████████████████████████████████████████████████████| 19/19 [00:00<?, ?it/s]
The config attributes {‘clip_sample’: False} were passed to PNDMScheduler, but are not expected and will be ignored. Please verify your scheduler_config.json configuration file.
Traceback (most recent call last):
File “”, line 1, in
File “C:\Users\win101user\AppData\Local\Env01\Lib\site-packages\diffusers\pipeline_utils.py”, line 455, in from_pretrained
load_method = getattr(class_obj, load_method_name)
TypeError: attribute name must be string, not ‘NoneType’

It looks like an error in the code but I don’t know much…

Any help would be appreciated, thanks !

These two lines work for me using diffusers 0.14.0, which is the latest version released today. Try updating your library (pip install diffusers==0.14.0), just in case.

In the instructions you mention, the only thing I can read about “access request” says:

[…]if you want to use pipelines that require an access request without having to be connected to the Hugging Face Hub, we recommend loading pipelines locally.

Note that you are giving a repository name to the from_pretrained() function, which means that it’s downloading the weights from the hub rather than your local filesystem (you should give it a path for that).

Thanks, updating diffusers made it work!