Цитата: Даниил от 31.03.2023, 18:07

Добрый день, проблема с LoRA делал четко по видео, но при запуске обучения появляется ошибка
[spoiler title="From https://github.com/bmaltais/kohya_ss
* branch master -> FETCH_HEAD
Already up to date.
Validating that requirements are satisfied.
All requirements satisfied.
Load CSS...
Running on local URL: http://127.0.0.1:7860

To create a public link, set `share=True` in `launch()`.
Folder 100_Audrey: 20 images found
Folder 100_Audrey: 2000 steps
max_train_steps = 1000
stop_text_encoder_training = 0
lr_warmup_steps = 100
accelerate launch --num_cpu_threads_per_process=2 "train_network.py" --pretrained_model_name_or_path="C:/Users/User/Documents/stable-diffusion/models/Stable-diffusion/deliberate_v2.safetensors" --train_data_dir="C:/Users/User/Pictures/train" --resolution=512,512 --output_dir="C:/Users/User/Pictures/train" --logging_dir="" --network_alpha="128" --save_model_as=safetensors --network_module=networks.lora --text_encoder_lr=5e-5 --unet_lr=0.0001 --network_dim=128 --output_name="Audrey" --lr_scheduler_num_cycles="1" --learning_rate="0.0001" --lr_scheduler="constant" --lr_warmup_steps="100" --train_batch_size="2" --max_train_steps="1000" --save_every_n_epochs="1" --mixed_precision="bf16" --save_precision="bf16" --cache_latents --optimizer_type="AdamW8bit" --max_data_loader_n_workers="1" --bucket_reso_steps=64 --xformers --bucket_no_upscale
prepare tokenizer
Traceback (most recent call last):
File "C:\Users\User\Documents\LoRA\train_network.py", line 711, in <module>
train(args)
File "C:\Users\User\Documents\LoRA\train_network.py", line 62, in train
tokenizer = train_util.load_tokenizer(args)
File "C:\Users\User\Documents\LoRA\library\train_util.py", line 2558, in load_tokenizer
tokenizer = CLIPTokenizer.from_pretrained(original_path)
File "C:\Users\User\Documents\LoRA\venv\lib\site-packages\transformers\tokenization_utils_base.py", line 1804, in from_pretrained
return cls._from_pretrained(
File "C:\Users\User\Documents\LoRA\venv\lib\site-packages\transformers\tokenization_utils_base.py", line 1852, in _from_pretrained
with open(tokenizer_config_file, encoding="utf-8") as tokenizer_config_handle:
OSError: [Errno 22] Invalid argument: 'tmp/.cache\\huggingface\\hub\\models--openai--clip-vit-large-patch14\\snapshots\\8d052a0f05efbaefbc9e8786ba291cfdf93e5bff\\tokenizer_config.json'
Traceback (most recent call last):
File "C:\Users\User\Documents\LoRA\python\lib\runpy.py", line 196, in _run_module_as_main
return _run_code(code, main_globals, None,
File "C:\Users\User\Documents\LoRA\python\lib\runpy.py", line 86, in _run_code
exec(code, run_globals)
File "C:\Users\User\Documents\LoRA\venv\Scripts\accelerate.exe\__main__.py", line 7, in <module>
File "C:\Users\User\Documents\LoRA\venv\lib\site-packages\accelerate\commands\accelerate_cli.py", line 45, in main
args.func(args)
File "C:\Users\User\Documents\LoRA\venv\lib\site-packages\accelerate\commands\launch.py", line 1104, in launch_command
simple_launcher(args)
File "C:\Users\User\Documents\LoRA\venv\lib\site-packages\accelerate\commands\launch.py", line 567, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['C:\\Users\\User\\Documents\\LoRA\\venv\\Scripts\\python.exe', 'train_network.py', '--pretrained_model_name_or_path=C:/Users/User/Documents/stable-diffusion/models/Stable-diffusion/deliberate_v2.safetensors', '--train_data_dir=C:/Users/User/Pictures/train', '--resolution=512,512', '--output_dir=C:/Users/User/Pictures/train', '--logging_dir=', '--network_alpha=128', '--save_model_as=safetensors', '--network_module=networks.lora', '--text_encoder_lr=5e-5', '--unet_lr=0.0001', '--network_dim=128', '--output_name=Audrey', '--lr_scheduler_num_cycles=1', '--learning_rate=0.0001', '--lr_scheduler=constant', '--lr_warmup_steps=100', '--train_batch_size=2', '--max_train_steps=1000', '--save_every_n_epochs=1', '--mixed_precision=bf16', '--save_precision=bf16', '--cache_latents', '--optimizer_type=AdamW8bit', '--max_data_loader_n_workers=1', '--bucket_reso_steps=64', '--xformers', '--bucket_no_upscale']' returned non-zero exit status 1."][/spoiler]
Видел что это обсуждалось, но решения не нашел, морочусь с этим 4 час, буду очень благодарен помощи более осведомлённых пользователей)