When running first time setup and specifying the vd_cache path, the code is not setup to handle the resulting file structure from the diffusers checkpoint download, producing the following error:
Traceback (most recent call last): File "/export/raid1/home/kneel027/fMRI-reconstruction-NSD/src/Reconstructions.py", line 199, in <module> vd_pipe.scheduler = UniPCMultistepScheduler.from_pretrained(vd_cache_dir, subfolder="scheduler") File "/export/raid1/home/kneel027/miniconda3/envs/mindeye/lib/python3.10/site-packages/diffusers/schedulers/scheduling_utils.py", line 139, in from_pretrained config, kwargs = cls.load_config( File "/export/raid1/home/kneel027/miniconda3/envs/mindeye/lib/python3.10/site-packages/diffusers/configuration_utils.py", line 320, in load_config raise EnvironmentError( OSError: Error no file named scheduler_config.json found in directory ../versatile_diffusion.
I had to correct the path used to load the scheduler from:
vd_pipe.scheduler = UniPCMultistepScheduler.from_pretrained(vd_cache_dir, subfolder="scheduler")
to the following to get it to load my downloaded checkpoint properly:
Hello,
When running first time setup and specifying the vd_cache path, the code is not setup to handle the resulting file structure from the diffusers checkpoint download, producing the following error:
Traceback (most recent call last): File "/export/raid1/home/kneel027/fMRI-reconstruction-NSD/src/Reconstructions.py", line 199, in <module> vd_pipe.scheduler = UniPCMultistepScheduler.from_pretrained(vd_cache_dir, subfolder="scheduler") File "/export/raid1/home/kneel027/miniconda3/envs/mindeye/lib/python3.10/site-packages/diffusers/schedulers/scheduling_utils.py", line 139, in from_pretrained config, kwargs = cls.load_config( File "/export/raid1/home/kneel027/miniconda3/envs/mindeye/lib/python3.10/site-packages/diffusers/configuration_utils.py", line 320, in load_config raise EnvironmentError( OSError: Error no file named scheduler_config.json found in directory ../versatile_diffusion.I had to correct the path used to load the scheduler from:
vd_pipe.scheduler = UniPCMultistepScheduler.from_pretrained(vd_cache_dir, subfolder="scheduler")to the following to get it to load my downloaded checkpoint properly:
vd_pipe.scheduler = UniPCMultistepScheduler.from_pretrained(vd_cache_dir + "/models--shi-labs--versatile-diffusion/snapshots/2926f8e11ea526b562cd592b099fcf9c2985d0b7", subfolder="scheduler")