LECO
1.0.0
확산 모델에서 개념을 지우기위한 저 순위 적응.
원래 저장소 : 확산 모델에서 개념을 지우는 것
그리고 프로젝트 페이지 : https://erasing.baulab.info/
(개념을 지우는 것뿐만 아니라 프롬프트와 로라 체중을 고안하여 개념을 강조하거나 교환합니다. 자세한 내용은 ConceptMod를 참조하십시오)
conda create -n leco python=3.10
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu118
pip install xformers
pip install -r requirements.txt최소한 8GB VRAM이 필요합니다.
python ./train_lora.py --config_file " ./examples/config.yaml " config.yaml :
prompts_file : " ./prompts.yaml "
pretrained_model :
name_or_path : " stabilityai/stable-diffusion-2-1 " # you can also use .ckpt or .safetensors models
v2 : true # true if model is v2.x
v_pred : true # true if model uses v-prediction
network :
type : " lierla " # or "c3lier"
rank : 4
alpha : 1.0
train :
precision : " bfloat16 "
noise_scheduler : " ddim " # or "ddpm", "lms", "euler_a"
iterations : 500
lr : 1e-4
optimizer : " AdamW "
lr_scheduler : " constant "
save :
name : " van_gogh "
path : " ./output "
per_steps : 200
precision : " bfloat16 "
logging :
use_wandb : false
verbose : false
other :
use_xformers : true prompts.yaml :
- target : " van gogh " # what word for erasing the positive concept from
positive : " van gogh " # concept to erase
unconditional : " " # word to take the difference from the positive concept
neutral : " " # starting point for conditioning the target
action : " erase " # erase or enhance
guidance_scale : 1.0
resolution : 512
dynamic_resolution : false
batch_size : 2자세한 내용은 예제 구성을 참조하십시오.
참고 : Float16을 사용할 수는 있지만 불안정하고 권장되지 않습니다. bfloat16 또는 float32를 사용하십시오.
Automatic1111의 WebUI에서 사전 가중량을 사용할 수 있습니다.
? huggingface : https://huggingface.co/p1atdev/leco
oil painting of van gogh by himself :
oil painting of van gogh by himself
Steps : 20, Sampler: Euler a, CFG scale: 7, Seed: 3870472781, Size: 512x512, Model hash: cc6cb27103, Model: v1-5-pruned-emaonly, Clip skip: 2, AddNet Enabled: True, AddNet Module 1: LoRA, AddNet Model 1: van_gogh_4_last(db68853d039b), AddNet Weight A 1: -1.0, AddNet Weight B 1: -1.0, Script: X/Y/Z plot, X Type: AddNet Weight 1, X Values: "-1, 0, 1", Version: v1.3.0 painting of scenery by monet 의 결과 :
painting of scenery by monet
Steps : 20, Sampler: Euler a, CFG scale: 7, Seed: 1284787312, Size: 512x512, Model hash: cc6cb27103, Model: v1-5-pruned-emaonly, Clip skip: 2, AddNet Enabled: True, AddNet Module 1: LoRA, AddNet Model 1: van_gogh_4_last(db68853d039b), AddNet Weight A 1: -1.0, AddNet Weight B 1: -1.0, Script: X/Y/Z plot, X Type: AddNet Weight 1, X Values: "-1, 0, 1", Version: v1.3.0 mona lisa with jewelry 의 결과 :
mona lisa with jewelry
Steps : 20, Sampler: Euler a, CFG scale: 7, Seed: 3630495347, Size: 512x512, Model hash: 832eb50c0c, Model: v2-1_768-ema-pruned, Clip skip: 2, AddNet Enabled: True, AddNet Module 1: LoRA, AddNet Model 1: mona_lisa2_last(393beb35c4b1), AddNet Weight A 1: -1.0, AddNet Weight B 1: -1.0, Script: X/Y/Z plot, X Type: AddNet Weight 1, X Values: "-1, 0, 1", Version: v1.3.0 photo of a cute cat 의 결과 :
photo of a cute cat
Steps : 20, Sampler: Euler a, CFG scale: 7, Seed: 900866192, Size: 512x512, Model hash: 832eb50c0c, Model: v2-1_768-ema-pruned, Clip skip: 2, AddNet Enabled: True, AddNet Module 1: LoRA, AddNet Model 1: mona_lisa2_last(393beb35c4b1), AddNet Weight A 1: -1.0, AddNet Weight B 1: -1.0, Script: X/Y/Z plot, X Type: AddNet Weight 1, X Values: "-1, 0, 1", Version: v1.3.0고양이 귀는 1.0 ~ 3.0 무게로 사용할 때 강제로 부착됩니다.
-1.0 ~ -3.0이면 고양이 귀는 나타나지 않습니다.
교육 설정 : 구성을 참조하십시오.
masterpiece, best quality, exceptional, best aesthetic, anime, 1girl, school uniform, upper body, smile
Negative prompt : worst quality, low quality, bad aesthetic, oldest, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username, blurry
Steps : 20, Sampler: Euler a, CFG scale: 7, Seed: 4103955758, Size: 512x512, Model hash: d38e779546, Model: wd-beta3-base-fp16, Clip skip: 2, Script: X/Y/Z plot, X Type: AddNet Weight 1, X Values: "0, 2, 3, 4", Version: v1.3.0교육 설정 : 구성을 참조하십시오.
"실생활, 인스 타 그램"으로 :
real life, instagram, masterpiece, best quality, exceptional, best aesthetic, 1girl, cat ears, blue hair, school uniform, upper body
Negative prompt : worst quality, low quality, bad aesthetic, oldest, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username, blurry
Steps : 20, Sampler: Euler a, CFG scale: 7, Seed: 757542759, Size: 768x768, Model hash: d38e779546, Model: wd-beta3-base-fp16, Clip skip: 2, AddNet Enabled: True, AddNet Module 1: LoRA, AddNet Model 1: unreal_6_many_prompts_200steps(fff5917285da), AddNet Weight A 1: -1.0, AddNet Weight B 1: -1.0, Script: X/Y/Z plot, X Type: AddNet Weight 1, X Values: "-1, 0, 1", Version: v1.3.0"실생활, 인스 타 그램"이 없으면 :
masterpiece, best quality, exceptional, best aesthetic,, 1girl, aqua eyes, baseball cap, blonde hair, closed mouth, earrings, green background, hat, hoop earrings, jewelry, looking at viewer, shirt, short hair, simple background, solo, upper body, yellow shirt,
Negative prompt : worst quality, low quality, bad aesthetic, oldest, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username, blurry
Steps : 20, Sampler: Euler a, CFG scale: 7, Seed: 2867636749, Size: 768x768, Model hash: d38e779546, Model: wd-beta3-base-fp16, Clip skip: 2, AddNet Enabled: True, AddNet Module 1: LoRA, AddNet Model 1: unreal_6_many_prompts_200steps(fff5917285da), AddNet Weight A 1: -1.0, AddNet Weight B 1: -1.0, Script: X/Y/Z plot, X Type: AddNet Weight 1, X Values: "-1, 0, 1", Version: v1.3.0 나는 깊은 영감을 받았으며 내 작업은 다음 프로젝트의 뛰어난 노력에 의존합니다. 이 프로젝트와 개발자에게 깊은 감사를 표하고 싶습니다.
https://github.com/rohitgandikota/erasing : 확산 모델에서 개념을 지우는 것
https://github.com/cloneofsimo/lora : 빠른 텍스트-이미지 확산 미세 조정을위한 저급 적응
https://github.com/kohya-ss/sd-scripts : 안정적인 확산을위한 교육, 생성 및 유틸리티 스크립트
https://github.com/ntc-ai/conceptmod : DSL을 사용하여 확산 모델에서 개념을 수정하십시오
