Closed mertayd0 closed 8 months ago
I tried to install bitsandbytes.cuda_setup.paths but it doesn't change anything.
Having the same issue here.
There seems to be some issue for some people, for me pip just refuse to install it correctly, it kept getting the old main file. I fixed it by manually extracting the wheel to the lib folder.
Same problem at my side.
sames here. Textual inversion works well, but Lora training has failed. I tried several thing such as installing an older version or installing the windows version bitSandBytes, cause it seems to me that error is related to bitSansBytes, but without success
My OS is win 10
--------------------------------------------Edit----------------------------------- AdamW8bit seems to be the source of the issue, if you change this by Adafactor for example it seems to work.
No really know what AdamW8bit is and its impact on training if you change that, so give it a try and go dive deep to doc
if youre on windows and want to use the adam8bit, change in the "requirements_windows_torch2.tx"t the "bitsandbytes==0.41.1 # no_verify" to "bitsandbytes-windows". Worked for me! it can be that you need to delete the bitsandbytes folder from the venv.
if youre on windows and want to use the adam8bit, change in the "requirements_windows_torch2.tx"t the "bitsandbytes==0.41.1 # no_verify" to "bitsandbytes-windows". Worked for me! it can be that you need to delete the bitsandbytes folder from the venv.
This is the solution! Took me a while to find this. Thank you so much!! Could the dev update the repo with this change?
prepare optimizer, data loader etc. Traceback (most recent call last): File "D:\SD\kohya_ss\library\train_util.py", line 3444, in get_optimizer import bitsandbytes as bnb File "D:\SD\kohya_ss\venv\lib\site-packages\bitsandbytes__init.py", line 6, in
from . import cuda_setup, utils, research
File "D:\SD\kohya_ss\venv\lib\site-packages\bitsandbytes\research__init__.py", line 1, in
from . import nn
File "D:\SD\kohya_ss\venv\lib\site-packages\bitsandbytes\research\nn\ init.py", line 1, in
from .modules import LinearFP8Mixed, LinearFP8Global
File "D:\SD\kohya_ss\venv\lib\site-packages\bitsandbytes\research\nn\modules.py", line 8, in
from bitsandbytes.optim import GlobalOptimManager
File "D:\SD\kohya_ss\venv\lib\site-packages\bitsandbytes\optim\ init__.py", line 6, in
from bitsandbytes.cextension import COMPILED_WITH_CUDA
File "D:\SD\kohya_ss\venv\lib\site-packages\bitsandbytes\cextension.py", line 5, in
from .cuda_setup.main import evaluate_cuda_setup
File "D:\SD\kohya_ss\venv\lib\site-packages\bitsandbytes\cuda_setup\main.py", line 21, in
from .paths import determine_cuda_runtime_lib_path
ModuleNotFoundError: No module named 'bitsandbytes.cuda_setup.paths'
During handling of the above exception, another exception occurred:
Traceback (most recent call last): File "D:\SD\kohya_ss\train_network.py", line 1012, in
trainer.train(args)
File "D:\SD\kohya_ss\train_network.py", line 342, in train
optimizer_name, optimizer_args, optimizer = train_util.get_optimizer(args, trainable_params)
File "D:\SD\kohya_ss\library\train_util.py", line 3446, in get_optimizer
raise ImportError("No bitsandbytes / bitsandbytesがインストールされていないようです")
ImportError: No bitsandbytes / bitsandbytesがインストールされていないようです
Traceback (most recent call last):
File "C:\Users\mb\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\mb\AppData\Local\Programs\Python\Python310\lib\runpy.py", line 86, in _run_code
exec(code, run_globals)
File "D:\SD\kohya_ss\venv\Scripts\accelerate.exe__main__.py", line 7, in
File "D:\SD\kohya_ss\venv\lib\site-packages\accelerate\commands\accelerate_cli.py", line 47, in main
args.func(args)
File "D:\SD\kohya_ss\venv\lib\site-packages\accelerate\commands\launch.py", line 986, in launch_command
simple_launcher(args)
File "D:\SD\kohya_ss\venv\lib\site-packages\accelerate\commands\launch.py", line 628, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['D:\SD\kohya_ss\venv\Scripts\python.exe', './train_network.py', '--enable_bucket', '--min_bucket_reso=256', '--max_bucket_reso=2048', '--pretrained_model_name_or_path=D:/SD/stable-diffusion-webui/models/Stable-diffusion/realisticVisionV60B1_v60B1VAE.safetensors', '--train_data_dir=C:/Users/mb/OneDrive/Desktop/Lora1/img', '--reg_data_dir=C:/Users/mb/OneDrive/Desktop/Lora1/reg', '--resolution=512,512', '--output_dir=C:/Users/mb/OneDrive/Desktop/Lora1/model', '--logging_dir=C:/Users/mb/OneDrive/Desktop/Lora1/log', '--network_alpha=32', '--save_model_as=safetensors', '--network_module=networks.lora', '--text_encoder_lr=2e-05', '--unet_lr=0.0002', '--network_dim=64', '--output_name=Car-A', '--lr_scheduler_num_cycles=5', '--no_half_vae', '--learning_rate=0.0002', '--lr_scheduler=constant', '--train_batch_size=2', '--max_train_steps=3750', '--save_every_n_epochs=1', '--mixed_precision=bf16', '--save_precision=bf16', '--cache_latents', '--cache_latents_to_disk', '--optimizer_type=AdamW8bit', '--max_grad_norm=1', '--max_data_loader_n_workers=0', '--bucket_reso_steps=64', '--xformers', '--bucket_no_upscale', '--noise_offset=0.0']' returned non-zero exit status 1.