뭔가 점점 발전되네 ㅋㅋ 나중엔 임베딩이나 하이퍼감마처럼 간편하게 나오려나

개 추

오

이미지 태그 파일은 따로 양식 없이 특징들 적어두면 되는거임? 퀄리티 태그같은건 빼고?

원래 설정에 학습률 없는거임...?
u-net 학습률하고 text encoder 학습률은 있는데 그냥 학습률은 없네...

모델 미리 넣어두는건 안되고 무조건 다운로드?

뭘로하던지 학습 결과물 자체는 비슷한가여?

구름에서 GPU칸이 안보이는데 도와주실수 있나요?https://arca.live/b/aiart/69008286/317720687#c_317720687

정규화 폴더(reg)엔 아무 것도 안 넣어도 잘 되나??

모델 다운받는데 civitai도 되나?

run 눌렀을때 실행중인지 끝난건지는 어케 구분해야함?
반응이 있는건지 없는건지 모르겠는데 그냥 기다려야하는건가;;

이거 학습할때마다 모델다운까지 처음부터 다시 해야함?
다운다받고 3. 데이터습득부터 하는데 학습이 몇초지나다가 끝나는데

궁금한게있는데 learning_rate 인수 넣는게 안보이는데 의도한거?
자동으로 2e-6으로 잡아주던데 kohya가 코랩에 기본값으로 적어논건 1e-4고, 0.000002이랑 0.0001은 차이 존나크지않나.

ModuleNotFoundError: No module named 'toml'
Traceback (most recent call last):
  File "/usr/local/bin/accelerate", line 8, in <module>
    sys.exit(main())
  File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/accelerate_cli.py", line 45, in main
    args.func(args)
  File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/launch.py", line 1104, in launch_command
    simple_launcher(args)
  File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/launch.py", line 567, in simple_launcher
    raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/usr/bin/python3', 'train_network.py', '--network_dim=128', '--network_alpha=128', '--network_module=networks.lora', '--learning_rate=0.0001', '--text_encoder_lr=5e-05', '--training_comment=training_comment', '--lr_scheduler=constant', '--pretrained_model_name_or_path=/workspace/repository/pre_trained_model/animefull-final-pruned.ckpt', '--caption_extension=.txt', '--train_data_dir=/workspace/repository/dreambooth/train_data', '--reg_data_dir=/workspace/repository/dreambooth/reg_data', '--output_dir=/workspace/repository/dreambooth/output', '--prior_loss_weight=1.0', '--output_name=Merumeru01', '--mixed_precision=fp16', '--save_precision=fp16', '--save_n_epoch_ratio=5', '--save_model_as=safetensors', '--resolution=512', '--train_batch_size=4', '--max_token_length=225', '--use_8bit_adam', '--max_train_epochs=10', '--gradient_accumulation_steps=1', '--clip_skip=2', '--logging_dir=/workspace/repository/dreambooth/logs', '--log_prefix=Merumeru01', '--shuffle_caption', '--enable_bucket', '--xformers', '--cache_latents']' returned non-zero exit status 1.
실행 부분에서 자꾸 이런 에러 뜨는데 이유가 뭘까
모듈 설정하는 파트랑 called process error 이부분을 모르겠음

2번에서 파일 다운 받고 업로드 하라는 사진에 보이는 창이 어딨는지 모르겠음 그냥 구름 실행하니까 검은색 코딩창만 뜨는데

와 코랩 로라는 보니까 머리빠개질거같던데 이거보니까 좀 쉬워보이네
뉴비 한번 도전해보러감 이런글 쓰러면 진짜 몇시간 걸릴텐데 이렇게 좋은글써줘서 넘고맙다....ㄹㅇ...

지금 구름에 WEB UI는설치해서 잘쓰고있는데 추가로 LoRa해보려고 하는데 한컨테이너에 같이 설치해도 되겠지?  
lora설치중  뭔가 오류생긴것같다  ModuleNotFoundError: No module named 'accelerate' 라고 뜨는데  왜그런걸까?  위명령문은 하나씩 다 실행시켜봤고,구글에 검색해보니 pip install accelerate해보라해서 해봐도 동일한 오류가 뜨더라

raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/usr/bin/python3', 'train_network.py', '--network_dim=128', '--network_alpha=128', '--network_module=networks.lora', '--learning_rate=0.0001', '--text_encoder_lr=5e-05', '--training_comment=training_comment', '--lr_scheduler=constant', '--pretrained_model_name_or_path=/workspace/repository/pre_trained_model/Chilloutmix.safetensors', '--caption_extension=.txt', '--train_data_dir=/workspace/repository/dreambooth/train_data', '--reg_data_dir=/workspace/repository/dreambooth/reg_data', '--output_dir=/workspace/repository/dreambooth/output', '--prior_loss_weight=1.0', '--output_name=Tw_dub', '--mixed_precision=fp16', '--save_precision=fp16', '--save_n_epoch_ratio=5', '--save_model_as=safetensors', '--resolution=512', '--train_batch_size=4', '--max_token_length=225', '--use_8bit_adam', '--max_train_epochs=10', '--gradient_accumulation_steps=1', '--clip_skip=2', '--logging_dir=/workspace/repository/dreambooth/logs', '--log_prefix=Tw_dub', '--shuffle_caption', '--enable_bucket', '--xformers', '--cache_latents']' returned non-zero exit status 1.

이거 무슨오류인지 이야기좀 부탁드립니다..

로라 학습 진행해보려니 에러가 나네요... 윗분이랑 비슷한듯 다른 문제인듯 싶은데 혹시 어떻게 해결할 수 있을까요..?

2023-04-24 05:39:21.042441: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-04-24 05:39:21.305225: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2023-04-24 05:39:22.174451: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory
2023-04-24 05:39:22.174605: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory
2023-04-24 05:39:22.174642: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
2023-04-24 05:39:25.137090: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-04-24 05:39:25.372250: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2023-04-24 05:39:26.280213: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory
2023-04-24 05:39:26.280370: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory
2023-04-24 05:39:26.280414: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
Traceback (most recent call last):
  File "train_network.py", line 18, in <module>
    import library.train_util as train_util
  File "/workspace/repository/kohya-trainer/library/train_util.py", line 63, in <module>
    import library.huggingface_util as huggingface_util
  File "/workspace/repository/kohya-trainer/library/huggingface_util.py", line 25, in <module>
    src: Union[str, Path, bytes, BinaryIO],
NameError: name 'BinaryIO' is not defined
Traceback (most recent call last):
  File "/usr/local/bin/accelerate", line 8, in <module>
    sys.exit(main())
  File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/accelerate_cli.py", line 45, in main
    args.func(args)
  File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/launch.py", line 1104, in launch_command
    simple_launcher(args)
  File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/launch.py", line 567, in simple_launcher
    raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/usr/bin/python3', 'train_network.py', '--network_dim=128', '--network_alpha=128', '--network_module=networks.lora', '--learning_rate=0.0001', '--text_encoder_lr=5e-05', '--training_comment=training_comment', '--lr_scheduler=constant', '--pretrained_model_name_or_path=/workspace/repository/pre_trained_model/ScrawlAbyss.safetensors', '--caption_extension=.txt', '--train_data_dir=/workspace/repository/dreambooth/train_data', '--reg_data_dir=/workspace/repository/dreambooth/reg_data', '--output_dir=/workspace/repository/dreambooth/output', '--prior_loss_weight=1.0', '--output_name=RRD', '--mixed_precision=fp16', '--save_precision=fp16', '--save_n_epoch_ratio=5', '--save_model_as=safetensors', '--resolution=512', '--train_batch_size=4', '--max_token_length=225', '--use_8bit_adam', '--max_train_epochs=10', '--gradient_accumulation_steps=1', '--clip_skip=2', '--logging_dir=/workspace/repository/dreambooth/logs', '--log_prefix=RRD', '--shuffle_caption', '--enable_bucket', '--xformers', '--cache_latents']' returned non-zero exit status 1.

센세..... 감사합니다