prithivMLmods commited on
Commit
32f51c4
·
verified ·
1 Parent(s): 5ce76c9

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +63 -0
README.md CHANGED
@@ -21,6 +21,69 @@ pipeline_tag: image-to-image
21
 
22
  <Gallery />
23
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
24
 
25
  ## Trigger words
26
 
 
21
 
22
  <Gallery />
23
 
24
+ ---
25
+
26
+ ## Sample Inference
27
+
28
+ | ex1 | ex2 |
29
+ |------|-------|
30
+ | ![Left Screenshot](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/3k76am1XiJJl4POgDI0cv.png) | ![Right Screenshot](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/5vl4iufT4NRWVNdJINtDL.png) |
31
+
32
+ ---
33
+
34
+ ## Parameter Settings
35
+
36
+ | Setting | Value |
37
+ | ------------------------ | ------------------------ |
38
+ | Module Type | Adapter |
39
+ | Base Model | FLUX.1 Kontext Dev - fp8 |
40
+ | Trigger Words | [photo content], transformed into a crochet plush doll, with visible yarn stitches, button eyes, and cozy handmade charm. |
41
+ | Image Processing Repeats | 50 |
42
+ | Epochs | 22 |
43
+ | Save Every N Epochs | 1 |
44
+
45
+ Labeling: DeepCaption-VLA-7B(natural language & English)
46
+
47
+ Total Images Used for Training : 28 Image Pairs (14 Start, 14 End)
48
+
49
+ Synthetic Result Node generated by NanoBanana from Google (Image Result Sets Dataset)
50
+
51
+
52
+ ## Training Parameters
53
+
54
+ | Setting | Value |
55
+ | --------------------------- | --------- |
56
+ | Seed | - |
57
+ | Clip Skip | - |
58
+ | Text Encoder LR | 0.00001 |
59
+ | UNet LR | 0.00005 |
60
+ | LR Scheduler | constant |
61
+ | Optimizer | AdamW8bit |
62
+ | Network Dimension | 64 |
63
+ | Network Alpha | 32 |
64
+ | Gradient Accumulation Steps | - |
65
+
66
+ ## Label Parameters
67
+
68
+ | Setting | Value |
69
+ | --------------- | ----- |
70
+ | Shuffle Caption | - |
71
+ | Keep N Tokens | - |
72
+
73
+ ## Advanced Parameters
74
+
75
+ | Setting | Value |
76
+ | ------------------------- | ----- |
77
+ | Noise Offset | 0.03 |
78
+ | Multires Noise Discount | 0.1 |
79
+ | Multires Noise Iterations | 10 |
80
+ | Conv Dimension | - |
81
+ | Conv Alpha | - |
82
+ | Batch Size | - |
83
+ | Steps | 2900 |
84
+ | Sampler | euler |
85
+
86
+ ---
87
 
88
  ## Trigger words
89