prithivMLmods commited on
Commit
7877bd0
1 Parent(s): 269ac8f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +19 -9
README.md CHANGED
@@ -24,7 +24,7 @@ license: mit
24
 
25
  <Gallery />
26
 
27
- ## Model description
28
 
29
  Image Processing Parameters
30
 
@@ -36,15 +36,25 @@ Image Processing Parameters
36
  | Network Alpha | 32 | Repeat & Steps | 30 & 4380|
37
  | Epoch | 20 | Save Every N Epochs | 1 |
38
 
39
- - Labeling: florence2-en(natural language & English)
40
-
41
- - Total Images Used for Training : 55 [ Hi-Res ]
42
-
43
- ## Best Dimensions
 
 
 
 
 
 
 
 
 
 
 
 
 
44
 
45
- - 1024 x 1024 (Default)
46
- - 768 x 1024
47
-
48
  ## Setting Up
49
  ```python
50
  import torch
 
24
 
25
  <Gallery />
26
 
27
+ ## Model description for super realism
28
 
29
  Image Processing Parameters
30
 
 
36
  | Network Alpha | 32 | Repeat & Steps | 30 & 4380|
37
  | Epoch | 20 | Save Every N Epochs | 1 |
38
 
39
+ ## Base model and other related model comparison
40
+
41
+ Comparison between the base model FLUX.1-dev and its adapter, a LoRA model tuned for super-realistic realism.
42
+
43
+ ![strangerzonehf/Flux-Super-Realism-LoRA](images/sz2.png)
44
+
45
+ However, it performs better in various aspects compared to its previous models, including face realism, ultra-realism, and others.
46
+
47
+ ![strangerzonehf/Flux-Super-Realism-LoRA](images/sz3.png)
48
+
49
+ ## Model Training Details
50
+
51
+ | Feature | Description |
52
+ |--------------------------------|-------------------------------------|
53
+ | **Labeling** | florence2-en (natural language & English) |
54
+ | **Total Images Used for Training** | 55 [Hi-Res] |
55
+ | **Best Dimensions** | - 1024 x 1024 (Default) |
56
+ | | - 768 x 1024 |
57
 
 
 
 
58
  ## Setting Up
59
  ```python
60
  import torch