Training Failure

[gt8101]

D:\Blender_ComfyUI\ComfyUI\custom_nodes\Lora-Training-in-Comfy/sd-scripts/train_network.py The following values were not passed toaccelerate launchand had defaults used instead: --num_processeswas set to a value of1 --num_machineswas set to a value of1 --mixed_precisionwas set to a value of'no' --dynamo_backendwas set to a value of'no' To avoid this warning pass in values for each of the problematic parameters or runaccelerate config. prepare tokenizer Traceback (most recent call last): File "D:\Blender_ComfyUI\ComfyUI\custom_nodes\Lora-Training-in-Comfy\sd-scripts\train_network.py", line 1012, in <module> trainer.train(args) File "D:\Blender_ComfyUI\ComfyUI\custom_nodes\Lora-Training-in-Comfy\sd-scripts\train_network.py", line 148, in train tokenizer = self.load_tokenizer(args) ^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\Blender_ComfyUI\ComfyUI\custom_nodes\Lora-Training-in-Comfy\sd-scripts\train_network.py", line 106, in load_tokenizer tokenizer = train_util.load_tokenizer(args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\Blender_ComfyUI\ComfyUI\custom_nodes\Lora-Training-in-Comfy\sd-scripts\library\train_util.py", line 3783, in load_tokenizer tokenizer = CLIPTokenizer.from_pretrained(original_path) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\Blender_ComfyUI\ComfyUI\.ext\Lib\site-packages\transformers\tokenization_utils_base.py", line 2032, in from_pretrained raise EnvironmentError( OSError: Can't load tokenizer for 'openai/clip-vit-large-patch14'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'openai/clip-vit-large-patch14' is the correct path to a directory containing all relevant files for a CLIPTokenizer tokenizer. Traceback (most recent call last): File "<frozen runpy>", line 198, in _run_module_as_main File "<frozen runpy>", line 88, in _run_code File "D:\Blender_ComfyUI\ComfyUI\.ext\Lib\site-packages\accelerate\commands\launch.py", line 1027, in <module> main() File "D:\Blender_ComfyUI\ComfyUI\.ext\Lib\site-packages\accelerate\commands\launch.py", line 1023, in main launch_command(args) File "D:\Blender_ComfyUI\ComfyUI\.ext\Lib\site-packages\accelerate\commands\launch.py", line 1017, in launch_command simple_launcher(args) File "D:\Blender_ComfyUI\ComfyUI\.ext\Lib\site-packages\accelerate\commands\launch.py", line 637, in simple_launcher raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd) subprocess.CalledProcessError: Command '['D:\\Blender_ComfyUI\\ComfyUI\\.ext\\python.exe', 'D:/Blender_ComfyUI/ComfyUI/custom_nodes/Lora-Training-in-Comfy/sd-scripts/train_network.py', '--enable_bucket', '--pretrained_model_name_or_path=D:\\Blender_ComfyUI\\ComfyUI\\models\\checkpoints\\epicphotogasm_zUniversal.safetensors', '--train_data_dir=D:/Blender_ComfyUI/ComfyUI/output', '--output_dir=D:\\Blender_ComfyUI\\ComfyUI\\models\\loras', '--logging_dir=./logs', '--log_prefix=realvix', '--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=realvix', '--train_batch_size=2', '--save_every_n_epochs=10', '--mixed_precision=fp16', '--save_precision=fp16', '--seed=17', '--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=2', '--optimizer_type=AdamW8bit', '--persistent_data_loader_workers', '--log_with=tensorboard', '--clip_skip=2', '--optimizer_type=AdamW8bit', '--persistent_data_loader_workers', '--log_with=tensorboard']' returned non-zero exit status 1. Train finished 2024-06-11 02:48:59,195- root:124- INFO- Prompt executed in 156.70 seconds

[Mintbeer96]

I encounter the same issue. I think the problem is that I updated the comfyui for sd3 and suddenly everything is broken. I am still debugging this and hopefully I can find something.

[Mintbeer96]

It turns out the transformer package and the kohya training script both updated. Just pull the lasted git and update your comfyui should work. The kohya script should go to sd-script.