AI绘画之训练自己的Lora模型

jqming

A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
Error no kernel image is available for execution on the device at line 167 in file D:\ai\tool\bitsandbytes\csrc\ops.cu
Traceback (most recent call last):
  File "C:\Users\liang\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\liang\AppData\Local\Programs\Python\Python310\lib\runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "D:\ZZZ\AI\lora-scripts\venv\Scripts\accelerate.exe\__main__.py", line 7, in <module>
  File "D:\ZZZ\AI\lora-scripts\venv\lib\site-packages\accelerate\commands\accelerate_cli.py", line 45, in main
    args.func(args)
  File "D:\ZZZ\AI\lora-scripts\venv\lib\site-packages\accelerate\commands\launch.py", line 1104, in launch_command
    simple_launcher(args)
  File "D:\ZZZ\AI\lora-scripts\venv\lib\site-packages\accelerate\commands\launch.py", line 567, in simple_launcher
    raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['D:\\ZZZ\\AI\\lora-scripts\\venv\\Scripts\\python.exe', './sd-scripts/train_network.py', '--enable_bucket', '--pretrained_model_name_or_path=./sd-models/model.ckpt', '--train_data_dir=./train/sucai', '--output_dir=./output', '--logging_dir=./logs', '--log_prefix=sucai', '--resolution=512,640', '--network_module=networks.lora', '--max_train_epochs=10', '--learning_rate=1e-4', '--unet_lr=1e-4', '--text_encoder_lr=1e-5', '--lr_scheduler=cosine_with_restarts', '--lr_warmup_steps=0', '--lr_scheduler_num_cycles=1', '--network_dim=32', '--network_alpha=32', '--output_name=sucai', '--train_batch_size=1', '--save_every_n_epochs=2', '--mixed_precision=fp16', '--save_precision=fp16', '--seed=1337', '--cache_latents', '--prior_loss_weight=1', '--max_token_length=225', '--caption_extension=.txt', '--save_model_as=safetensors', '--min_bucket_reso=256', '--max_bucket_reso=1024', '--keep_tokens=0', '--xformers', '--shuffle_caption', '--clip_skip=2', '--optimizer_type=AdamW8bit']' returned non-zero exit status 1.
Train finished


是什么问题

zuolian

devilshy 发表于 2023-4-16 17:47
解决了No module named 'xformers'，又报错啦Error caught was: No module named 'triton'

试过直 ...

这个问题不用管它，

onda1478963

运行完了 "install-cn.ps1"
SDwebUI在哪打开？

herons

感谢大神的指导很有用

wiyifan

很详细的教程

kaka123

[Dataset 0]
loading image sizes.
100%|████████████████████████████████████████████████████████████████████████████████| 57/57 [00:00<00:00, 2597.94it/s]
make buckets
number of images (including repeats) / 各bucketの画像枚数（繰り返し回数を含む）
bucket 0: resolution (512, 512), count: 171
mean ar error (without repeats): 0.0
prepare accelerator
Traceback (most recent call last):
  File "I:\pro\lora-scripts\sd-scripts\train_network.py", line 773, in <module>
    train(args)
  File "I:\pro\lora-scripts\sd-scripts\train_network.py", line 139, in train
    accelerator, unwrap_model = train_util.prepare_accelerator(args)
  File "I:\pro\lora-scripts\sd-scripts\library\train_util.py", line 2847, in prepare_accelerator
    accelerator = Accelerator(
  File "I:\pro\lora-scripts\venv\lib\site-packages\accelerate\accelerator.py", line 355, in __init__
    raise ValueError(err.format(mode="fp16", requirement="a GPU"))
ValueError: fp16 mixed precision requires a GPU
Traceback (most recent call last):
  File "C:\Users\weir\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\weir\AppData\Local\Programs\Python\Python310\lib\runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "I:\pro\lora-scripts\venv\Scripts\accelerate.exe\__main__.py", line 7, in <module>
  File "I:\pro\lora-scripts\venv\lib\site-packages\accelerate\commands\accelerate_cli.py", line 45, in main
    args.func(args)

2605088505

A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
训练提示的，能看看嘛，真是呜呜呜

oyqm

一脸懵逼的先回复再看，十几年没有摸电脑了

2715410008

raise ValueError(err.format(mode="fp16", requirement="a GPU"))
ValueError: fp16 mixed precision requires a GPU

楼主这个报错怎么搞呀

mmll

想要玩的６，要学的东西太多了啊