Closed BlinkerHigh closed 8 months ago
+1 same error,hepl!!tks!
I keep getting the same error: "raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd) subprocess.CalledProcessError: Command '['d:\kohya_ss\venv\Scripts\python.exe', './sdxl_train_network.py', '--enable_bucket', '--min_bucket_reso=256', '--max_bucket_reso=2048', '--pretrained_model_name_or_path=stabilityai/stable-diffusion-xl-base-1.0', '--train_data_dir=D:/Art/_ML datasets/Style/Destination\img', '--resolution=1024,1024', '--output_dir=D:/Art/_ML datasets/Style/Destination\model', '--logging_dir=D:/Art/_ML datasets/Style/Destination\log', '--network_alpha=1', '--training_comment=trigger: STYLE', '--save_model_as=safetensors', '--network_module=networks.lora', '--text_encoder_lr=0.00012', '--unet_lr=0.0012', '--network_dim=128', '--gradient_accumulation_steps=2', '--output_name=STYLE', '--lr_scheduler_num_cycles=1', '--no_half_vae', '--learning_rate=0.0012', '--lr_scheduler=constant', '--train_batch_size=1', '--max_train_steps=58200', '--save_every_n_epochs=1', '--mixed_precision=bf16', '--save_precision=bf16', '--caption_extension=.txt', '--cache_latents', '--cache_latents_to_disk', '--optimizer_type=AdamW', '--optimizer_args', 'scale_parameter=False', 'relative_step=False', 'warmup_init=False', '--max_data_loader_n_workers=0', '--bucket_reso_steps=64', '--gradient_checkpointing', '--xformers', '--bucket_no_upscale', '--noise_offset=0.0']' returned non-zero exit status 1."
Any idea how to solve this issue?
I already took care of it,Someone commented elsewhere that running Setup can resolve the bitsandbytes issues.You can have a try.#1692
I keep getting the same error: "raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd) subprocess.CalledProcessError: Command '['d:\kohya_ss\venv\Scripts\python.exe', './sdxl_train_network.py', '--enable_bucket', '--min_bucket_reso=256', '--max_bucket_reso=2048', '--pretrained_model_name_or_path=stabilityai/stable-diffusion-xl-base-1.0', '--train_data_dir=D:/Art/_ML datasets/Style/Destination\img', '--resolution=1024,1024', '--output_dir=D:/Art/_ML datasets/Style/Destination\model', '--logging_dir=D:/Art/_ML datasets/Style/Destination\log', '--network_alpha=1', '--training_comment=trigger: STYLE', '--save_model_as=safetensors', '--network_module=networks.lora', '--text_encoder_lr=0.00012', '--unet_lr=0.0012', '--network_dim=128', '--gradient_accumulation_steps=2', '--output_name=STYLE', '--lr_scheduler_num_cycles=1', '--no_half_vae', '--learning_rate=0.0012', '--lr_scheduler=constant', '--train_batch_size=1', '--max_train_steps=58200', '--save_every_n_epochs=1', '--mixed_precision=bf16', '--save_precision=bf16', '--caption_extension=.txt', '--cache_latents', '--cache_latents_to_disk', '--optimizer_type=AdamW', '--optimizer_args', 'scale_parameter=False', 'relative_step=False', 'warmup_init=False', '--max_data_loader_n_workers=0', '--bucket_reso_steps=64', '--gradient_checkpointing', '--xformers', '--bucket_no_upscale', '--noise_offset=0.0']' returned non-zero exit status 1." Any idea how to solve this issue?
I already took care of it,Someone commented elsewhere that running Setup can resolve the bitsandbytes issues.You can have a try.#1692
What do you mean by running setup?
I just reinstalled Kohya SS and I still get the same error...
I just reinstalled Kohya SS and I still get the same error...
Start PowerShell with administrator rights: Set-ExecutionPolicy RemoteSigned, then close the window.
Non-administrators start PowerShell: Switch to the folder directory (cd X:\X\kohya_ss), without switching if you are on drive C. a. Create a venv virtual environment file: python -m venv venv b. Start the virtual environment: venv\Scripts\activate.ps1
activate activate.bat: .\venv\Scripts\activate.bat
Uninstall bitsandbytespip: pip uninstall bitsandbytes Y
Run the following command to install bitsandbytes: python -m pip install bitsandbytes --prefer-binary --extra-index-url=https://jllllll.github.io/bitsandbytes-windows-webui
5. bitsandbytes: python -m pip install bitsandbytes --prefer-binary --extra-index-url=https://jllllll.github.io/bitsandbytes-windows-webui
Does not work I am still getting the following error: E:\kohya_ss\venv\lib\site-packages\torch\utils\checkpoint.py:31: UserWarning: None of the inputs have requires_grad=True. Gradients will be None warnings.warn("None of the inputs have requires_grad=True. Gradients will be None")
- bitsandbytes: python -m pip install bitsandbytes --prefer-binary --extra-index-url=https://jllllll.github.io/bitsandbytes-windows-webui
Does not work I am still getting the following error: E:\kohya_ss\venv\lib\site-packages\torch\utils\checkpoint.py:31: UserWarning: None of the inputs have requires_grad=True. Gradients will be None warnings.warn("None of the inputs have requires_grad=True. Gradients will be None")
Hello,
I've tried every solution but it doesn't work... Still the same problem... Textual Inversion or Lora, training doesn't work. I really need help, someone... :) Thanks on advance.
`11:17:43-666063 INFO Version: v22.6.0
11:17:43-670830 INFO nVidia toolkit detected 11:17:44-971307 INFO Torch 2.1.2+cu118 11:17:44-988147 INFO Torch backend: nVidia CUDA 11.8 cuDNN 8700 11:17:44-989112 INFO Torch detected GPU: NVIDIA GeForce RTX 3070 Ti VRAM 8192 Arch (8, 6) Cores 48 11:17:44-990149 INFO Verifying modules installation status from requirements_windows_torch2.txt... 11:17:44-992149 INFO Installing package: torch==2.1.2+cu118 torchvision==0.16.2+cu118 torchaudio==2.1.2+cu118 --index-url https://download.pytorch.org/whl/cu118 11:17:48-096708 INFO Installing package: xformers==0.0.23.post1+cu118 --index-url https://download.pytorch.org/whl/cu118 11:17:49-891744 INFO Verifying modules installation status from requirements.txt... 11:17:49-894655 WARNING Package wrong version: gradio 3.36.1 required 3.44.0 11:17:49-895655 INFO Installing package: gradio==3.44.0 11:17:59-816991 INFO headless: False 11:17:59-818991 INFO Load CSS... Running on local URL: http://127.0.0.1:7860
To create a public link, set share=True
in launch()
.
11:21:51-279574 INFO Start training TI...
11:21:51-296412 INFO Valid image folder names found in: E:/Images/AI/Training/Characters/Roji_Clover/images
11:21:51-297413 INFO Folder 25_Roji_Clover: 3750 steps
11:21:51-298412 INFO max_train_steps = 18750
11:21:51-299412 INFO stop_text_encoder_training = 0
11:21:51-300413 INFO lr_warmup_steps = 1875
11:21:51-301412 INFO Saving training config to
E:/Images/AI/Training/Characters/Roji_Clover/model\RojiClover_20240207-112151.json...
11:21:51-551664 INFO accelerate launch --num_cpu_threads_per_process=2 "./train_textual_inversion.py"
--enable_bucket --min_bucket_reso=256 --max_bucket_reso=2048
--pretrained_model_name_or_path="runwayml/stable-diffusion-v1-5"
--train_data_dir="E:/Images/AI/Training/Characters/Roji_Clover/images" --resolution="768,5768"
--output_dir="E:/Images/AI/Training/Characters/Roji_Clover/model"
--logging_dir="E:/Images/AI/Training/Characters/Roji_Clover/logs" --save_model_as=safetensors
--output_name="RojiClover" --lr_scheduler_num_cycles="5" --max_data_loader_n_workers="0"
--learning_rate="1e-05" --lr_scheduler="cosine" --lr_warmup_steps="1875" --train_batch_size="1"
--max_train_steps="18750" --save_every_n_epochs="1" --mixed_precision="fp16"
--save_precision="fp16" --cache_latents --optimizer_type="AdamW8bit"
--max_data_loader_n_workers="0" --bucket_reso_steps=64 --xformers --bucket_no_upscale
--noise_offset=0.0 --token_string="woman" --init_word="" --num_vectors_per_token=14
A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
prepare tokenizer
prepare accelerator
loading model for process 0/1
load Diffusers pretrained models: runwayml/stable-diffusion-v1-5
Loading pipeline components...: 100%|████████████████████████████████████████████████████| 5/5 [00:00<00:00, 12.70it/s]
You have disabled the safety checker for <class 'diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline'> by passing safety_checker=None
. Ensure that you abide to the conditions of the Stable Diffusion license and do not expose unfiltered results in services or applications open to the public. Both the diffusers team and Hugging Face strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling it only for use-cases that involve analyzing network behavior or auditing its results. For more information, please have a look at https://github.com/huggingface/diffusers/pull/254 .
UNet2DConditionModel: 64, 8, 768, False, False
U-Net converted to original U-Net
Traceback (most recent call last):
File "D:\Programmes 2\Logiciels\kohya_ss-master\train_textual_inversion.py", line 797, in
I keep getting the same error: "raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd) subprocess.CalledProcessError: Command '['d:\kohya_ss\venv\Scripts\python.exe', './sdxl_train_network.py', '--enable_bucket', '--min_bucket_reso=256', '--max_bucket_reso=2048', '--pretrained_model_name_or_path=stabilityai/stable-diffusion-xl-base-1.0', '--train_data_dir=D:/Art/_ML datasets/Style/Destination\img', '--resolution=1024,1024', '--output_dir=D:/Art/_ML datasets/Style/Destination\model', '--logging_dir=D:/Art/_ML datasets/Style/Destination\log', '--network_alpha=1', '--training_comment=trigger: STYLE', '--save_model_as=safetensors', '--network_module=networks.lora', '--text_encoder_lr=0.00012', '--unet_lr=0.0012', '--network_dim=128', '--gradient_accumulation_steps=2', '--output_name=STYLE', '--lr_scheduler_num_cycles=1', '--no_half_vae', '--learning_rate=0.0012', '--lr_scheduler=constant', '--train_batch_size=1', '--max_train_steps=58200', '--save_every_n_epochs=1', '--mixed_precision=bf16', '--save_precision=bf16', '--caption_extension=.txt', '--cache_latents', '--cache_latents_to_disk', '--optimizer_type=AdamW', '--optimizer_args', 'scale_parameter=False', 'relative_step=False', 'warmup_init=False', '--max_data_loader_n_workers=0', '--bucket_reso_steps=64', '--gradient_checkpointing', '--xformers', '--bucket_no_upscale', '--noise_offset=0.0']' returned non-zero exit status 1."
Any idea how to solve this issue?