Modified PonyDiffusionV6XL for LoRA training
This model may improve the quality of LoRA.
not suitable for artstyle LoRAs and photorealistic style
このモデルはLoRAの品質を改善するかもしれません。
画風LoRAと実写風には向いていません。
Training information:
Finetuned PonyDiffusionV6XL by repeating the training and merging a LoRA two times with sd-scripts.
sd-scriptsで二回loraの学習とマージを繰り返すことでファインチューンされました。
DO NOT USE THESE PARAMETERS FOR LORA!
Training parameters(This is not a recommended setting):
Resolution:1024x1024
Bucketing: enabled,min512,max2048
LR scheduler: cosine_with_restarts
Optimizer: Lion
Keep tokens:0
Network dim/alpha: 8/4
Conv dim/alpha:1/0.25
Gradient checkpointing:true
Grad acc steps:1
Batch size: 2
Epochs:1
Unet learning rate:0.00015
Shuffle caption:true
Seed:42
LR warmup steps:100
Dataset information:
Dataset size: 4157
Training script(source code)
sd-scripts( https://github.com/kohya-ss/sd-scripts )
Notice/おしらせ:
This model is licensed under the Fair AI Public License 1.0-SD.
If you make modify this model, you must share both your changes and the original license.
You are not permitted to run inference of this model on websites or applications allowing any form of monetization (paid inference, faster tiers, etc.). This applies to any derivative models or model merges.
このモデルはFair AI Public License 1.0-SDでライセンスされています。
このモデルを改変する場合、変更点および元のライセンスを明記する必要があります。
このモデルの生成サービスでの商用利用を禁止します。これは継承先にも適用されます。
Description
FAQ
Comments (2)
Can you upload a json with the Settings for Lora Training?
I tried to use your serttings but always get an error with my rtx 3090
Traceback (most recent call last):
File "C:\100sd\kohya_sdxl\sd-scripts\sdxl_train_network.py", line 185, in <module>
trainer.train(args)
File "C:\100sd\kohya_sdxl\sd-scripts\train_network.py", line 198, in train
train_dataset_group = config_util.generate_dataset_group_by_blueprint(blueprint.dataset_group)
File "C:\100sd\kohya_sdxl\sd-scripts\library\config_util.py", line 482, in generate_dataset_group_by_blueprint
dataset = dataset_klass(subsets=subsets, **asdict(dataset_blueprint.params))
File "C:\100sd\kohya_sdxl\sd-scripts\library\train_util.py", line 1451, in init
max(resolution) <= max_bucket_reso
AssertionError: max_bucket_reso must be equal or greater than resolution / max_bucket_resoは最大解像度より小さくできません。解像度を小さくするかmin_bucket_resoを大きくしてください
Traceback (most recent call last):
File "C:\Users\cosci\AppData\Local\Programs\Python\Python310\lib\runpy.py", line 196, in runmodule_as_main
return runcode(code, main_globals, None,
File "C:\Users\cosci\AppData\Local\Programs\Python\Python310\lib\runpy.py", line 86, in runcode
exec(code, run_globals)
File "C:\100sd\kohya_sdxl\venv\Scripts\accelerate.exe\__main__.py", line 7, in <module>
File "C:\100sd\kohya_sdxl\venv\lib\site-packages\accelerate\commands\accelerate_cli.py", line 47, in main
args.func(args)
File "C:\100sd\kohya_sdxl\venv\lib\site-packages\accelerate\commands\launch.py", line 1017, in launch_command
simple_launcher(args)
File "C:\100sd\kohya_sdxl\venv\lib\site-packages\accelerate\commands\launch.py", line 637, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['C:\\100sd\\kohya_sdxl\\venv\\Scripts\\python.exe', 'C:/100sd/kohya_sdxl/sd-scripts/sdxl_train_network.py', '--config_file', './outputs/tmpfilelora.toml', '--optimizer_args', 'weight_decay=0.01', 'd_coef=1', 'use_bias_correction=True', 'safeguard_warmup=False', 'betas=0.9,0.99']' returned non-zero exit status 1.
Try set resolution=1024,min_bucket_reso=512,max_bucket_reso=2048,enable_bucket
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.