search

fudan-generative-vision / hallo

Hallo: Hierarchical Audio-Driven Visual Synthesis for Portrait Image Animation
https://fudan-generative-vision.github.io/hallo/
MIT License
9.52k stars 1.3k forks source link

Diffusers Issue #205

Open ldavis9000aws opened 2 weeks ago

ldavis9000aws commented 2 weeks ago

I've went through the install and am receiving the following error when I run this command for inference. The "pretrained_models" directory was created and populated without an issue prior to running inference.

python scripts/inference.py --source_image examples/reference_images/1.jpg --driving_audio examples/driving_audios/1.wav

Traceback (most recent call last): File "/home/ubuntu/hallo/scripts/inference.py", line 35, in from diffusers import AutoencoderKL, DDIMScheduler File "/home/ubuntu/miniconda3/envs/hallo/lib/python3.10/site-packages/diffusers/init.py", line 5, in from .utils import ( File "/home/ubuntu/miniconda3/envs/hallo/lib/python3.10/site-packages/diffusers/utils/init.py", line 38, in from .dynamic_modules_utils import get_class_from_dynamic_module File "/home/ubuntu/miniconda3/envs/hallo/lib/python3.10/site-packages/diffusers/utils/dynamic_modules_utils.py", line 28, in from huggingface_hub import cached_download, hf_hub_download, model_info ImportError: cannot import name 'cached_download' from 'huggingface_hub' (/home/ubuntu/miniconda3/envs/hallo/lib/python3.10/site-packages/huggingface_hub/init.py)

Here is my CUBA driver version:

(hallo) ubuntu@ip-10-0-12-195:~/hallo$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Mon_Apr__3_17:16:06_PDT_2023
Cuda compilation tools, release 12.1, V12.1.105
Build cuda_12.1.r12.1/compiler.32688072_0
(hallo) ubuntu@ip-10-0-12-195:~/hallo$

piplist.txt

mudassirkhan19 commented 2 weeks ago

Downgrading hf hub should work: https://github.com/easydiffusion/easydiffusion/issues/1851#issuecomment-2445843321