Closed lzlzlz1999 closed 11 months ago
This is may be because hugging face is not accept for you because of the network issue. You should use a VPN to download the pretrain model from hugging face to your server or local computer, then change the path of the encoder to the model you have downloaded.
Yes, I thought of the same solution.
(ldm) lz@manager-Precision-7920-Tower:~/Documents/InST$ python main.py --base configs/stable-diffusion/v1-finetune.yaml -t --actual_resume ./models/sd/sd-v1-4.ckpt -n log1_shuimo --gpus 0, --data_root /home/lz/Documents/InST/style Global seed set to 23 Running on GPUs 0, Loading model from ./models/sd/sd-v1-4.ckpt LatentDiffusion: Running in eps-prediction mode DiffusionWrapper has 859.52 M params. making attention of type 'vanilla' with 512 in_channels Working with z of shape (1, 4, 32, 32) = 4096 dimensions. making attention of type 'vanilla' with 512 in_channels Traceback (most recent call last): File "main.py", line 582, in
model = load_model_from_config(config, opt.actual_resume)
File "main.py", line 29, in load_model_from_config
model = instantiate_from_config(config.model)
File "/home/lz/Documents/InST/ldm/util.py", line 85, in instantiate_from_config
return get_obj_from_str(config["target"])(config.get("params", dict()), kwargs)
File "/home/lz/Documents/InST/ldm/models/diffusion/ddpm.py", line 477, in init
self.instantiate_cond_stage(cond_stage_config)
File "/home/lz/Documents/InST/ldm/models/diffusion/ddpm.py", line 561, in instantiate_cond_stage
model = instantiate_from_config(config)
File "/home/lz/Documents/InST/ldm/util.py", line 85, in instantiate_from_config
return get_obj_from_str(config["target"])(config.get("params", dict()), kwargs)
File "/home/lz/Documents/InST/ldm/modules/encoders/modules.py", line 166, in init
self.tokenizer = CLIPTokenizer.from_pretrained(version)
File "/home/lz/anaconda3/envs/ldm/lib/python3.8/site-packages/transformers/tokenization_utils_base.py", line 1764, in from_pretrained
raise EnvironmentError(
OSError: Can't load tokenizer for 'openai/clip-vit-large-patch14'. 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 'openai/clip-vit-large-patch14' is the correct path to a directory containing all relevant files for a CLIPTokenizer tokenizer.
During handling of the above exception, another exception occurred:
Traceback (most recent call last): File "main.py", line 795, in
if trainer.global_rank == 0:
NameError: name 'trainer' is not defined