image
上次分享過基本的stable diffusion入門使用
可以參考這篇:
https://monkeyjhong.pixnet.net/blog/post/340168266

今天要更進一步的用
chilloutmix模型創造出以假亂真的照片
模型下載網址:

https://civitai.com/models/6424/chilloutmix

 

抓下來的model放進stable diffusion模型資料夾
路徑 stable-diffusion-webui\models\Stable-diffusion

image

 

運行stable diffusion
已經run起來的話點那個藍色銜尾蛇刷新

image

image

同樣上方模型下載網址往下滾動會有別人訓練好的模型參考


image

在茫茫人海中一眼被這個假志孝吸住
image

照這邊提供的參數填入後生成
出來吧 神奇寶貝志孝~~~

?????!!!!!


image

內心一驚難道志孝是靠整容整成至孝的嗎??!
檢查後發現model沒切換過去chilloutmix

倒帶~
再照這個參數

()是用來加強某個詞的效果的,多套幾層(),效果會一步步加強,也可以用冒號加數字的格式,比如(word:1.5)來修改權限;

 


a woman holding a brown bag and smiling at the camera with her hands on her chest and her arm around her neck
Negative prompt: EasyNegative, paintings, sketches, (worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, ((monochrome)), ((grayscale)), skin spots, acnes, skin blemishes, age spot, glans,extra fingers,fewer fingers,strange fingers,bad hand,
ENSD: 31337, Size: 512x768, Seed: 2113451480, Steps: 20, Sampler: DPM++ 2S a Karras, CFG scale: 7.5, Clip skip: 2, Model hash: fc2511737a, Hires steps: 15, Hires upscale: 1.5, AddNet Enabled: True, AddNet Model 1: songzhixiao(5e2560b6c310), Hires upscaler: Latent, AddNet Module 1: LoRA, AddNet Weight A 1: 0.8, AddNet Weight B 1: 0.8, Denoising strength: 0.5

image


順利長出以假亂真的照片拉
雖然根治孝有點落差


image

 

再換一組參數
1girl,beautiful vintage color, instagram (photorealistic, high resolution:1.4), ((puffy eyes)), looking at viewer, , full body (8k, RAW photo, best quality, masterpiece:1.2), (realistic, photo-realistic:1.37),(sharp focus:1.2), professional lighting, photon mapping, radiosity, physically-based rendering, (pale skin:1.2),,(small breasts:1.2),, looking at viewer, (middle hair:1.2), portrait, purple eyes, (sliver hair:1.2),bangs, (simple background:1.4), solo, upper body, white background,realistic,(masterpiece:1.4),(best quality:1.4),(shiny skin),fashi-girl,makeup,(closed mouth:1.4),(skinny,,shy, :1.3),(white Turtleneck sweater:1.2), ulzzang-6500-v1.1, pureerosface_v1, <lora: japaneseDollLikeness_v10:0.3>, <lora:yh:0.4>, <lora:bc:0.30>, <lora:xh:0.4>,
Negative prompt: ng_deepnegative_v1_75t,EasyNegative, paintings, sketches, (worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, ((monochrome)), ((grayscale)), skin spots, acnes, skin blemishes, age spot, (outdoor:1.6), manboobs, backlight,(ugly:1.331), (duplicate:1.331), (morbid:1.21), (mutilated:1.21), (tranny:1.331), mutated hands, (poorly drawn hands:1.331), blurry, (bad anatomy:1.8), (bad proportions:1.331), extra limbs, (disfigured:1.331), (more than 2 nipples:1.331), (missing arms:1.331), (extra legs:1.331), (fused fingers:1.61051), (too many fingers:1.61051), (unclear eyes:1.331), bad hands, missing fingers, extra digit, (futa:1.1), bad body, glans,long neck,extra fingers, extra limbs, extra arms, extra legs,flower,glasses,teeth
Seed: 354364784, Steps: 34, Sampler: DPM++ SDE Karras, CFG scale: 7

image

image

 

 

image

image

 

https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Command-Line-Arguments-and-Settings

image

image

image

 

升級python到3.10 
把新的python路徑放進webui-user.bat
砍掉venv資料夾
重跑webui-user.bat

 

image

 

安裝Kohya-ss Additional Networks

image

 

 

image

apply 成功後會看到長出additional networks頁籤 

image

 

image

image

 

arrow
arrow
    文章標籤
    chilloutmix stable idffusion
    全站熱搜

    MonkeyJ 發表在 痞客邦 留言(0) 人氣()