AI绘画之训练自己的Lora模型

抹茶味牛子

1354402178 发表于 2023-2-26 09:40
accelerate : 无法将“accelerate”项识别为 cmdlet、函数、脚本文件或可运行程序的名称。请检查名称的拼写 ...

看常见错误排除

抹茶味牛子

timlee0719 发表于 2023-2-25 23:51
請問大佬和版主, 我也是同樣的問題, 不管切換哪個模型(Pretrained model )都是同一個問題

loading text  ...

train文件删掉--use_8bit_adam

timlee0719

抹茶味牛子 发表于 2023-2-26 16:01
train文件删掉--use_8bit_adam

開始運作了, 感謝版主!!

k16576

Fatal error in launcher: Unable to create process using '"D:\deepfacelab\lora??\lora-scripts\venv\Scripts\python.exe"  "D:\deepfacelab\lora\lora-scripts\venv\Scripts\accelerate.exe" launch --num_cpu_threads_per_process=8 ./sd-scripts/train_network.py --enable_bucket --pretrained_model_name_or_path=./sd-models/model.ckpt --train_data_dir=./train/en --output_dir=./output --logging_dir=./logs --resolution=512,512 --network_module=networks.lora --max_train_epochs=20 --learning_rate=1e-4 --unet_lr=1e-4 --text_encoder_lr=1e-5 --lr_scheduler=cosine_with_restarts --lr_warmup_steps=0 --network_dim=32 --network_alpha=32 --output_name=en --train_batch_size=1 --save_every_n_epochs=2 --mixed_precision=fp16 --save_precision=fp16 --seed=1337 --cache_latents --clip_skip=2 --prior_loss_weight=1 --max_token_length=225 --caption_extension=.txt --save_model_as=safetensors --min_bucket_reso=256 --max_bucket_reso=1024 --xformers --shuffle_caption ': ???????????

Train finished
这个是什么情况呀，有没有大神知道怎么解决

rexay

accelerate : The term 'accelerate' is not recognized as the name of a cmdlet, function, script file, or operable progra
m. Check the spelling of the name, or if a path was included, verify that the path is correct and try again.
At D:\BaiduNetdiskDownload\lora训练环境\lora-scripts\train.ps1:72 char:1
+ accelerate launch --num_cpu_threads_per_process=8 "./sd-scripts/train ...
+ ~~~~~~~~~~
    + CategoryInfo          : ObjectNotFound: (accelerate:String) [], CommandNotFoundException
    + FullyQualifiedErrorId : CommandNotFoundException

Train finished



请问出现这种情况怎么办呀

抹茶味牛子

k16576 发表于 2023-2-27 04:38
Fatal error in launcher: Unable to create process using '"D:\deepfacelab\lora??\lora-scripts\venv\Sc ...

加群贴报错图

抹茶味牛子

rexay 发表于 2023-2-27 19:06
accelerate : The term 'accelerate' is not recognized as the name of a cmdlet, function, script file, ...

强制更新一下试试

paco90

抹茶味牛子 发表于 2023-2-25 15:04
SD1.5是所有db丹的丹底，就是万物始祖，其他注入chilloutmax等模型也都是基于这个练出来的，所以这个模型 ...

涨知识了，谢谢大佬

zwbtc

炼丹开始 ，然后就这样了 。。

prepare tokenizer
Traceback (most recent call last):
  File "E:\kohya_ss\train_network.py", line 507, in <module>
    train(args)
  File "E:\kohya_ss\train_network.py", line 56, in train
    tokenizer = train_util.load_tokenizer(args)
  File "E:\kohya_ss\library\train_util.py", line 1802, in load_tokenizer
    tokenizer = CLIPTokenizer.from_pretrained(TOKENIZER_PATH)
  File "E:\kohya_ss\venv\lib\site-packages\transformers\tokenization_utils_base.py", line 1804, in from_pretrained
    return cls._from_pretrained(
  File "E:\kohya_ss\venv\lib\site-packages\transformers\tokenization_utils_base.py", line 1959, in _from_pretrained
    tokenizer = cls(*init_inputs, **init_kwargs)
  File "E:\kohya_ss\venv\lib\site-packages\transformers\models\clip\tokenization_clip.py", line 323, in __init__
    self.encoder = json.load(vocab_handle)
  File "C:\Users\ui\AppData\Local\Programs\Python\Python310\lib\json\__init__.py", line 293, in load
    return loads(fp.read(),
  File "C:\Users\ui\AppData\Local\Programs\Python\Python310\lib\json\__init__.py", line 346, in loads
    return _default_decoder.decode(s)
  File "C:\Users\ui\AppData\Local\Programs\Python\Python310\lib\json\decoder.py", line 337, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "C:\Users\ui\AppData\Local\Programs\Python\Python310\lib\json\decoder.py", line 353, in raw_decode
    obj, end = self.scan_once(s, idx)
json.decoder.JSONDecodeError: Unterminated string starting at: line 1 column 214903 (char 214902)
Traceback (most recent call last):
  File "C:\Users\ui\AppData\Local\Programs\Python\Python310\lib\runpy.py", line 196, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "C:\Users\ui\AppData\Local\Programs\Python\Python310\lib\runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "E:\kohya_ss\venv\Scripts\accelerate.exe\__main__.py", line 7, in <module>
  File "E:\kohya_ss\venv\lib\site-packages\accelerate\commands\accelerate_cli.py", line 45, in main
    args.func(args)
  File "E:\kohya_ss\venv\lib\site-packages\accelerate\commands\launch.py", line 1104, in launch_command
    simple_launcher(args)
  File "E:\kohya_ss\venv\lib\site-packages\accelerate\commands\launch.py", line 567, in simple_launcher
    raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['E:\\kohya_ss\\venv\\Scripts\\python.exe', 'train_network.py', '--pretrained_model_name_or_path=D:/miaoshouai_stable-diffusion-webui-cn_230205/stable-diffusion-webui-cn/models/Stable-diffusion/v1-5-pruned-emaonly.ckpt', '--train_data_dir=E:/xunlian/image', '--resolution=512,512', '--output_dir=D:/miaoshouai_stable-diffusion-webui-cn_230205/stable-diffusion-webui-cn/models/Lora', '--logging_dir=E:/xunlian/log', '--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=rognarok1', '--lr_scheduler_num_cycles=1', '--learning_rate=0.0001', '--lr_scheduler=constant', '--train_batch_size=1', '--max_train_steps=4700', '--save_every_n_epochs=1', '--mixed_precision=fp16', '--save_precision=fp16', '--seed=1234', '--caption_extension=.txt', '--cache_latents', '--optimizer_type=AdamW8bit', '--max_data_loader_n_workers=1', '--clip_skip=2', '--bucket_reso_steps=64', '--mem_eff_attn', '--gradient_checkpointing', '--xformers', '--use_8bit_adam', '--bucket_no_upscale']' returned non-zero exit status 1.

zwbtc

zwbtc 发表于 2023-3-1 16:57
炼丹开始 ，然后就这样了 。。

prepare tokenizer

折腾了1天还是这样。。求助