syaimu commited on
Commit
5582a29
1 Parent(s): 01f788f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +17 -1
README.md CHANGED
@@ -30,7 +30,23 @@ Negative prompt : (worst quality:1.6),(low quality:1.4),(normal quality:1.2),low
30
  <img src="https://files.catbox.moe/70gi4h.jpg" width="1700" height="">
31
 
32
 
33
- # <Marge: 7th anime XL BASE :0.7:lbw=1:0,1,1,1,0.9,0.7,0.5,0.3,0.1,0,0,0,0.1,0.3,0.5,0.7,0.9,1,1,1>.
34
 
35
  <img src="https://i.imgur.com/5K87RDg.jpeg" width="1700" height="">
36
  <img src="https://i.imgur.com/Qseddy9.jpeg" width="1700" height="">
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
30
  <img src="https://files.catbox.moe/70gi4h.jpg" width="1700" height="">
31
 
32
 
33
+ # <lora: 7th anime XL BASE :0.255:lbw=1:0,1,1,1,0.9,0.7,0.5,0.3,0.1,0,0,0,0.1,0.3,0.5,0.7,0.9,1,1,1>.
34
 
35
  <img src="https://i.imgur.com/5K87RDg.jpeg" width="1700" height="">
36
  <img src="https://i.imgur.com/Qseddy9.jpeg" width="1700" height="">
37
+
38
+ # <Marge: The Recipe :0.7>
39
+ 1.
40
+ Merge Animagine 3.0 and 3.1 using a base alpha of 0.49, and merge layers from IN00 to OUT11 at 0.82.
41
+ 2.
42
+ Train a model on sd_xl_base_1.0_0.9vae.safetensors with ~4.6 million images on A100x4, at a learning rate of 1e-5, for ~2 epochs, and then, for compatibility with Animagine models' CLIP modules, further train it with a dataset of 164 AI-generated images to refine CLIP and Unet, using PRODIGY, Initial D at 1e-06, D Coefficient at 0.9, and a batch size of 4 for 1500 steps.
43
+ 3.
44
+ Merge 1. and 2. using two sets of coefficients:
45
+ ・Set 1: 0.2, 0.6, 0.8, 0.9, 0.0, 0.8, 0.4, 1.0, 0.7, 0.9, 0.3, 0.1, 0.1, 0.5, 0.6, 0.0, 1.0, 0.6, 0.5, 0.5
46
+ ・Set 2: 0.9, 0.8, 0.6, 0.3, 0.9, 0.1, 0.4, 0.7, 0.4, 0.6, 0.2, 0.3, 0.0, 0.8, 0.3, 0.7, 0.7, 0.8, 0.2, 0.3.
47
+ 4.
48
+ Merge Set 1 and Set 2 using a base alpha of 0.79 and merge layers from IN00 to OUT11 at 0.73 to create Set 3.
49
+ 5.
50
+ Train a LoRA based on Set 3 with a curated dataset of 12,018 AI-generated images, Lion optimizer, batch size of 4, gradient accumulation steps of 16, and learning rate 3e-5 for 4 epochs. This model is then blended into Set 3 itself at a strength of 0.2, resulting in the creation of 7th anime B.
51
+ 6.
52
+ Train another LoRA based on 7th anime B with the same dataset as described in step 2. but with Lion optimizer, batch size of 4, lr_scheduler_num_cycles at 5, and learning rate 1e-5 for 80 epochs. This model is then blended into 7th anime B itself at a strength of 0.366, finally resulting in the creation of 7th anime A.