File size: 1,771 Bytes
8334f8f
 
170d1c3
 
 
8334f8f
9c7b9bd
e977363
d3a5e51
5676afe
 
200e5e5
 
ee652e9
 
 
 
 
 
 
 
9c7b9bd
50fe16d
9c7b9bd
 
 
 
 
3867547
8f745bf
c529f6a
50fe16d
 
 
 
 
 
 
8f745bf
538642d
8f745bf
 
 
 
 
 
5d63d53
538642d
 
 
 
52c6d56
538642d
 
 
5d63d53
c529f6a
5d63d53
 
 
 
 
 
c529f6a
 
 
 
 
52c6d56
c529f6a
 
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
---
license: openrail
tags:
- stable diffusion
- controlnet
---

Can be used in [sd-webui-controlnet](https://github.com/Mikubill/sd-webui-controlnet). Put the yaml file with model because it is sd2.1 v_prediction model.

Model is the difference from the original checkpoint created by [extract_controlnet_diff.py](https://github.com/Mikubill/sd-webui-controlnet/blob/main/extract_controlnet_diff.py).

These are not the final version and need to be adjusted to be more accurate. Someone please do this.

# wd15_controlnet_edge_depth_pose.ckpt
+ target model: https://huggingface.co/waifu-diffusion/wd-1-5-beta2
+ dataset: 20k 1girl images
+ epoch: 5
+ batch size: 8
+ lr: 5e-5
+ amp:bfloat16

# diff_control_wd15beta2_canny.safetensors
**updated in 2023/03/05.**
+ target model: https://huggingface.co/waifu-diffusion/wd-1-5-beta2
+ dataset: 11k 1girl images
+ epoch: 2
+ batch size: 16
+ lr: 5e-5
+ amp:bfloat16

and resumed on:

+ dataset: 60k 1girl images
+ epoch: 2
+ batch size: 16
+ lr: 1e-5
+ amp:bfloat16

# diff_control_wd15beta2_pose.safetensors
**updated in 2023/03/06.**
+ target model: https://huggingface.co/waifu-diffusion/wd-1-5-beta2
+ dataset: 14k umamusume images. **Therefore, the model may have a bias**
+ epoch: 5 (The “sudden convergence phenomenon” occurred at the fourth epoch)
+ batch size: 16
+ lr: 5e-5
+ amp:bfloat16

and resumed on:

+ dataset: 52k 1girl images
+ epoch: 2
+ batch size: 16
+ lr: 1e-5
+ amp:bfloat16

# diff_control_wd15beta2_depth.safetensors
**updated in 2023/03/06.**
+ target model: https://huggingface.co/waifu-diffusion/wd-1-5-beta2
+ dataset: 11k 1girl images
+ epoch: 2
+ batch size: 16
+ lr: 5e-5
+ amp:bfloat16

and resumed on:

+ dataset: 60k 1girl images
+ epoch: 1
+ batch size: 16
+ lr: 1e-5
+ amp:bfloat16