search

dvlab-research / Video-P2P

Video-P2P: Video Editing with Cross-attention Control
https://video-p2p.github.io/
386 stars 25 forks source link

Error #10

Closed xh9998 closed 1 year ago

xh9998 commented 1 year ago

OSError: Can't load tokenizer for '/cpfs01/user/lixiaohui/code/Video-P2P/outputs/rabbitjump'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'Video-P2P/outputs/rabbitjump' is the correct path to a directory containing all relevant files for a CLIPTokenizer tokenizer.

ShaoTengLiu commented 1 year ago

Could you share your running scripts?

The default saving location is outputs/rabbit-jump. Did you correctly save your weights in outputs/rabbitjump?

xh9998 commented 1 year ago

yes,I do. I change it to jump rabbit. I try another yaml but still got this error. Let me show you:

python run_videop2p.py --config="configs/man-motor-p2p.yaml" Traceback (most recent call last): File "/cpfs01/user/lixiaohui/code/Video-P2P/run_videop2p.py", line 664, in main(**OmegaConf.load(args.config), fast=args.fast) File "/cpfs01/user/lixiaohui/code/Video-P2P/run_videop2p.py", line 75, in main tokenizer = CLIPTokenizer.from_pretrained(pretrained_model_path, subfolder="tokenizer") File "/cpfs01/user/lixiaohui/.conda/envs/TAV/lib/python3.9/site-packages/transformers/tokenization_utils_base.py", line 1825, in from_pretrained raise EnvironmentError( OSError: Can't load tokenizer for './outputs/man-motor'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure './outputs/man-motor' is the correct path to a directory containing all relevant files for a CLIPTokenizer tokenizer.

xh9998 commented 1 year ago
image

I have got ddim_latent-500.pt, isnt that enough?

ShaoTengLiu commented 1 year ago

If you run the first stage (run_tuning.py) properly. There should be outputs/rabbit-jump/tokenizer.

Please check whether this line is executed: line.

xh9998 commented 1 year ago

ok, thank you. It turns out that the question lies in tuneavideo. TypeError: The keyword fps is no longer supported. Use duration(in ms) instead, e.g. fps=50 == duration=20 (1000 * 1/50). Afer I fix this, I get the right tokenizer.

ShaoTengLiu commented 1 year ago

Thanks for pointing out.