Sasulee commited on
Commit
3e47f91
1 Parent(s): 248ff5c

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +65 -0
README.md ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: bsd
3
+ language:
4
+ - ab
5
+ metrics:
6
+ - accuracy
7
+ - bleu
8
+ - cer
9
+ library_name: keras
10
+ tags:
11
+ - art
12
+ ---
13
+
14
+ # Anime Lineart / Manga-like (线稿/線画/マンガ風/漫画风) Style
15
+
16
+ <!-- Provide a quick summary of what the model is/does. -->
17
+
18
+ Update information
19
+ V3.0 updated. Fine-tuned LoRA to improve the effects of generating characters with complex body limbs and backgrounds. The overall styling is more toward manga style rather than simple lineart. A lower CFG scale can make lines thinner.
20
+
21
+ V3.0更新。对LoRA进行了微调,改善生成人物肢体和复杂背景的效果。整体更趋向漫画风而非纯线稿。低CFG scale可以让线条变细。
22
+
23
+ Recommend settings
24
+ Model: Anything V4.5
25
+
26
+ VAE: Orangemix (the same with NAI)
27
+
28
+ LoRA Strength: 1
29
+
30
+ Sampler: DPM++ 2M Karras
31
+
32
+ Sampling steps: 20
33
+
34
+ CFG: 7
35
+
36
+ Negative embedding: EasyNegative、badhandv4
37
+
38
+ Highres fix is also recommended.
39
+
40
+ Note: all the LoRA names used in sample images are my local name, you need to change them to your saved LoRA filename!
41
+
42
+ 基模:Anything V4.5
43
+
44
+ VAE:Orangemix(同NAI)
45
+
46
+ LoRA强度:1
47
+
48
+ 采样器:DPM++ 2M Karras
49
+
50
+ 采样步数:20
51
+
52
+ CFG:7
53
+
54
+ 负面embedding:EasyNegative、badhandv4
55
+
56
+ 推荐使用高清修复。
57
+
58
+ 注意:样图中的LoRA名称是我本地文件名,要改成你保存的文件名!
59
+
60
+ Difference
61
+ Here's a comparision of V2.0 and V3.0 (locally called V3 and V4 on my machine - V2 on my machine is not published):
62
+
63
+ V2.0和V3.0的对比(在我的本地机器上命名为V3和V4——本地的V2我并未发布)
64
+
65
+ [![Alt text](https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/38354d83-7523-4334-c3de-1a3a493e5a00/width=525/38354d83-7523-4334-c3de-1a3a493e5a00)]