prithivMLmods commited on
Commit
e3cb35a
·
verified ·
1 Parent(s): d4daac2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +47 -1
README.md CHANGED
@@ -22,12 +22,58 @@ widget:
22
  text: 'Glass Cube, A brilliant gold nugget is suspended within a crystal-clear cube-shaped block. The nugget has a naturally irregular surface that shimmers under the lighting, casting faint golden reflections on the inner glass walls. Below the nugget, the words Au Gold 79 are printed in bold black text. The background is stark white, allowing the rich, metallic hue of the gold to stand out vividly. A subtle shadow from the cube falls to the right, adding dimension to the composition.'
23
  base_model: black-forest-labs/FLUX.1-dev
24
  instance_prompt: Glass Cube
25
-
26
  ---
 
 
 
27
  # Flux-Material-Glass-Cube
28
 
29
  <Gallery />
30
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
31
 
32
  ## Trigger words
33
 
 
22
  text: 'Glass Cube, A brilliant gold nugget is suspended within a crystal-clear cube-shaped block. The nugget has a naturally irregular surface that shimmers under the lighting, casting faint golden reflections on the inner glass walls. Below the nugget, the words Au Gold 79 are printed in bold black text. The background is stark white, allowing the rich, metallic hue of the gold to stand out vividly. A subtle shadow from the cube falls to the right, adding dimension to the composition.'
23
  base_model: black-forest-labs/FLUX.1-dev
24
  instance_prompt: Glass Cube
 
25
  ---
26
+
27
+ ![1.png](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/9tZ4Pb9WViboSswJjBxjg.png)
28
+
29
  # Flux-Material-Glass-Cube
30
 
31
  <Gallery />
32
 
33
+ ---
34
+
35
+ # Model description for Flux-Material-Glass-Cube
36
+
37
+ Image Processing Parameters
38
+
39
+ | Parameter | Value | Parameter | Value |
40
+ |---------------------------|--------|---------------------------|--------|
41
+ | LR Scheduler | constant | Noise Offset | 0.03 |
42
+ | Optimizer | AdamW | Multires Noise Discount | 0.1 |
43
+ | Network Dim | 64 | Multires Noise Iterations | 10 |
44
+ | Network Alpha | 32 | Repeat & Steps | 18 & 2100 |
45
+ | Epoch | 14 | Save Every N Epochs | 1 |
46
+
47
+ Labeling: florence2-en(natural language & English)
48
+
49
+ Total Images Used for Training : 17
50
+
51
+ ## Best Dimensions & Inference
52
+
53
+ | **Dimensions** | **Aspect Ratio** | **Recommendation** |
54
+ |-----------------|------------------|---------------------------|
55
+ | 1280 x 832 | 3:2 | Best |
56
+ | 1024 x 1024 | 1:1 | Default |
57
+
58
+ ### Inference Range
59
+
60
+ - **Recommended Inference Steps:** 30-35
61
+
62
+ ## Setting Up
63
+ ```python
64
+ import torch
65
+ from pipelines import DiffusionPipeline
66
+
67
+ base_model = "black-forest-labs/FLUX.1-dev"
68
+ pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
69
+
70
+ lora_repo = "strangerzonehf/Flux-Material-Glass-Cube"
71
+ trigger_word = "Glass Cube"
72
+ pipe.load_lora_weights(lora_repo)
73
+
74
+ device = torch.device("cuda")
75
+ pipe.to(device)
76
+ ```
77
 
78
  ## Trigger words
79