Open BusyGettingBusy opened 1 month ago
Scorpio-mfg
commented
1 month ago same error
ComfyUI_windows_portable\custom_nodes\Lora-Training-in-Comfy\sd-scripts\train_network.py': [Errno 2] No such file or directory
Traceback (most recent call last):
xtrash
commented
3 weeks ago Hi. This is because the "ComfyUI" folder is not on the path to the missing file. You should have "ComfyUI" between "\ComfyUI_windows_portable" and "\custom_nodes\". So -> \ComfyUI_windows_portable\ComfyUI\custom_nodes... The easiest way is to start ComfyUI from the "ComfyUI" forlder, not from the "ComfyUI_windows_portable" folder.
REM Switch python session to virtual environment if needed : call ComfyUI\venv\Scripts\activate.bat echo "Venv 03 Activated"
cd ComfyUI
..\python_embeded\python.exe -s main.py --windows-standalone-build --port 8030 --cuda-device 0 pause
Thats how I solved this problem, but you can edit the "train.py" file (in Lora-Training-in-Comfy) to change the differents "progpath" variable affectation. I think there is something to do with the "progpath = os.getcwd()" function .
progpath = os.getcwd()
BusyGettingBusy
commented
3 weeks ago Thank you @xtrash for your response. However, I updated "run_nvidia_gpu.bat" to precisely what you pasted and I am getting the same error still.
C:\Users\balco\AI_local_stuff\ComfyUI_windows_portable\ComfyUI\custom_nodes\Lora-Training-in-Comfy/sd-scripts/train_network.py
The following values were not passed to `accelerate launch` and had defaults used instead:
`--num_processes` was set to a value of `1`
`--num_machines` was set to a value of `1`
`--mixed_precision` was set to a value of `'no'`
`--dynamo_backend` was set to a value of `'no'`
To avoid this warning pass in values for each of the problematic parameters or run `accelerate config`.
Traceback (most recent call last):
File "C:\Users\balco\AI_local_stuff\ComfyUI_windows_portable\ComfyUI\custom_nodes\Lora-Training-in-Comfy\sd-scripts\train_network.py", line 11, in <module>
import toml
ModuleNotFoundError: No module named 'toml'
Traceback (most recent call last):
File "C:\Users\balco\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\balco\AppData\Local\Programs\Python\Python310\lib\runpy.py", line 86, in _run_code
exec(code, run_globals)
File "C:\Users\balco\AppData\Local\Programs\Python\Python310\lib\site-packages\accelerate\commands\launch.py", line 989, in <module>
main()
File "C:\Users\balco\AppData\Local\Programs\Python\Python310\lib\site-packages\accelerate\commands\launch.py", line 985, in main
launch_command(args)
File "C:\Users\balco\AppData\Local\Programs\Python\Python310\lib\site-packages\accelerate\commands\launch.py", line 979, in launch_command
simple_launcher(args)
File "C:\Users\balco\AppData\Local\Programs\Python\Python310\lib\site-packages\accelerate\commands\launch.py", line 628, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['C:\\Users\\balco\\AppData\\Local\\Programs\\Python\\Python310\\python.exe', 'C:/Users/balco/AI_local_stuff/ComfyUI_windows_portable/ComfyUI/custom_nodes/Lora-Training-in-Comfy/sd-scripts/train_network.py', '--enable_bucket', '--pretrained_model_name_or_path=C:\\Users\\balco\\AI_local_stuff\\ComfyUI_windows_portable\\ComfyUI\\models\\checkpoints\\v1-5-pruned.safetensors', '--train_data_dir=C:/Users/balco/AI_local_stuff/SD LORA training/prepare training data output/img', '--output_dir=models/loras', '--logging_dir=./logs', '--log_prefix=maimytj-v0.1-comfyui', '--resolution=512,512', '--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=maimytj-v0.1-comfyui', '--train_batch_size=1', '--save_every_n_epochs=1', '--mixed_precision=fp16', '--save_precision=fp16', '--seed=12', '--cache_latents', '--prior_loss_weight=1', '--max_token_length=225', '--caption_extension=.txt', '--save_model_as=safetensors', '--min_bucket_reso=256', '--max_bucket_reso=1584', '--keep_tokens=0', '--xformers', '--shuffle_caption', '--clip_skip=1', '--optimizer_type=AdamW8bit', '--persistent_data_loader_workers', '--log_with=tensorboard']' returned non-zero exit status 1.
Train finished
Prompt executed in 2.06 secondsClosed by accident.
xtrash
commented
3 weeks ago Hi BusyGettingBusy. I was answering Scorpio-mfg about the path issue.
By the way, I remember struggling with the " No module named 'toml' " error too. I don't know exactly how I solved this, but here is what I did :
.\python_embeded\python.exe -s main.py --windows-standalone-build --port 8030 --cuda-device 0
checked my Venv config file (pyvenv.cfg) that actually looks like this :
home = W:\ComfyUI_03\ComfyUI_windows_portable\python_embeded implementation = CPython version_info = 3.10.11.final.0 virtualenv = 20.26.2 include-system-site-packages = false base-prefix = W:\ComfyUI_03\ComfyUI_windows_portable\python_embeded base-exec-prefix = W:\ComfyUI_03\ComfyUI_windows_portable\python_embeded base-executable = W:\ComfyUI_03\ComfyUI_windows_portable\python_embeded\python.exe
Updated all the libraries on python (in the Virtual Environment) with :
pip freeze > requirements.txt
then changed "==" for ">=" in the requirements.txt text file I get, and then updated everything with :
pip install -r requirements.txt --upgrade
To exactly know the last version, I use this trick trying to update torch (or torchvision and torchaudio) with torch=+cu121 that will return an error with the last version number :
pip install torch==+cu121 --index-url https://download.pytorch.org/whl/cu121
The answer will look like :
ERROR: Could not find a version that satisfies the requirement torch==+cu121 (from versions: 2.1.0+cu121, 2.1.1+cu121, 2.1.2+cu121, 2.2.0+cu121, 2.2.1+cu121, 2.2.2+cu121, 2.3.0+cu121, 2.3.1+cu121)
The last version number is at the end of the error : 2.3.1+cu121
now you can reinstall torch with the correct version number :
pip install torch==2.3.1+cu121 --index-url https://download.pytorch.org/whl/cu121
After that, I did the same with torchvision and torchaudio.
I don't remember why I decided to update everything, but it worked for me. I must admit that I was surprised that Lora-Training-in-Comfy finally works with everything up to date and not sticking to the requirements file of this custom node.
Hello, I have seen similar errors in the other issues posted, but they were all different. None of these solutions solved my problem.
I ran the advanced LORA trainer and I keep getting the following error. I see the error "[Errno 2] No such file or directory". However, I checked and this is the path to that file. I also checked that the path to the images folder is correct (that folder contains the folder that contains the images).
By the way, I also ran in the non-advanced version, and it gave me a similar error, but with traceback saying this. However, I am focused on fixing the advanced one.