File size: 3,117 Bytes
7e56521 ffc6f9f 7e56521 ffc6f9f dae86c4 65301a0 dae86c4 44063eb dae86c4 8a1a6a9 dae86c4 a172bf8 dae86c4 a172bf8 dae86c4 36c7be8 f65a354 6b4926d 8a207d4 89484b0 aa386ed 25b3dee dae86c4 9f5840a dae86c4 01bfc41 f051f83 a172bf8 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 |
---
tags:
- text-to-image
- stable-diffusion
- diffusion-models-class
- dreambooth-hackathon
- wildcard
---
# If you want to use this style , the prompt: wgz style
**I trianed this weight by using dreambooth**
There are three models in the library, V1, V2, and beta
V1:WOLP.ckpt
V2:wgz v2.ckpt
BEAT :wgz beta.ckpt
Let me introduce these versions ,
V1 is the earliest. use 512 resolution for training,
Advantages: the generation logic is good, text2img is used well
Disadvantages: There will be less screen details and less imitation of strokes. The whites of the eyes turn red easily.
V2 adjusted the training set and used 768 resolution training
Advantages: more details, better facial portrayal
Disadvantages: Eyeballs tend to generate green, and the logic of text generation is not good, resulting in poor usability of text2img.
BETA: A version of the training map that is relatively complex has poor logic, but sometimes satisfactory results can be obtained in the img2 img. Try if you want!
**v2**
Next, I will show the nice output of V2, which is made by using the img2img. I will also give the configuration I use.
**parameters;**
Prompt :wgz style,portrait of a beautiful women highly detailed,perfect femine face,Classical oil painting,by masamune shirow, by William-tae Kim
Negative prompt: old,men,two poeple,Three dimensional facial features,deep shadows on the five senses,tall nose,Eye shadow,[High bridge], [narrow nose],Deep orbital fossa
Steps: 20, Sampler: DPM++ 2S a Karras, CFG scale: 9.5, Denoising strength: 0.51, Mask blur: 4
I hope you guys enjoy this model and create works that satisfied
**BETA**
Below I will show the nice output of a beta. I like its abstract color swatches, but the production rate is not high
I hope you guys enjoy this model and create works that satisfied
Below is the introduction about V1,and some nice output by using V1
**v1**
--------------------------
Training with the Elden ring Style ckpt （SD1.5）
# link：[eldenring-ckpt](https://huggingface.co/nitrosocke/elden-ring-diffusion)
# Training parameters：
steps： 12500 stpes
Learning rate：1e-6
int images number：about 100
class image number ：200
**There are some samples:**
PS:I try hard to achieve the painting style I want by using stable diffusion. I have tried a lot, with failures and successes. With my current technology, I can hardly find progress now. So this model will not be updated in a short time. If you have ideas or want to improve together, you can contact me,happy new year guys.
Telegram: https://t.me/hhhyxnh
QQ:1602821649
** More pictures in the image folder
thank you. |