prithivMLmods commited on
Commit
79b2995
·
verified ·
1 Parent(s): 29591e7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +42 -0
README.md CHANGED
@@ -27,12 +27,52 @@ base_model: black-forest-labs/FLUX.1-dev
27
  instance_prompt: Icon Kit
28
  license: creativeml-openrail-m
29
  ---
 
30
 
31
  <Gallery />
32
 
33
  # Model description for Flux-Icon-Kit-LoRA
34
 
 
35
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
36
  ## Trigger words
37
 
38
  You should use `Icon Kit` to trigger the image generation.
@@ -42,3 +82,5 @@ You should use `Icon Kit` to trigger the image generation.
42
  Weights for this model are available in Safetensors format.
43
 
44
  [Download](/strangerzonehf/Flux-Icon-Kit-LoRA/tree/main) them in the Files & versions tab.
 
 
 
27
  instance_prompt: Icon Kit
28
  license: creativeml-openrail-m
29
  ---
30
+ ![icon pack.png](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/G_QXbbtYT1YFflykerj-m.png)
31
 
32
  <Gallery />
33
 
34
  # Model description for Flux-Icon-Kit-LoRA
35
 
36
+ Image Processing Parameters
37
 
38
+ | Parameter | Value | Parameter | Value |
39
+ |---------------------------|--------|---------------------------|--------|
40
+ | LR Scheduler | constant | Noise Offset | 0.03 |
41
+ | Optimizer | AdamW | Multires Noise Discount | 0.1 |
42
+ | Network Dim | 64 | Multires Noise Iterations | 10 |
43
+ | Network Alpha | 32 | Repeat & Steps | 25 & 3100 |
44
+ | Epoch | 20 | Save Every N Epochs | 1 |
45
+
46
+ Labeling: florence2-en(natural language & English)
47
+
48
+ Total Images Used for Training : 40 [ Raw 8-bit ]
49
+
50
+ ## Best Dimensions & Inference
51
+
52
+ | **Dimensions** | **Aspect Ratio** | **Recommendation** |
53
+ |-----------------|------------------|---------------------------|
54
+ | 1280 x 832 | 3:2 | Best |
55
+ | 1024 x 1024 | 1:1 | Default |
56
+
57
+ ### Inference Range
58
+
59
+ - **Recommended Inference Steps:** 30–35
60
+
61
+ ## Setting Up
62
+ ```python
63
+ import torch
64
+ from pipelines import DiffusionPipeline
65
+
66
+ base_model = "black-forest-labs/FLUX.1-dev"
67
+ pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
68
+
69
+ lora_repo = "strangerzonehf/Flux-Icon-Kit-LoRA"
70
+ trigger_word = "Icon Kit"
71
+ pipe.load_lora_weights(lora_repo)
72
+
73
+ device = torch.device("cuda")
74
+ pipe.to(device)
75
+ ```
76
  ## Trigger words
77
 
78
  You should use `Icon Kit` to trigger the image generation.
 
82
  Weights for this model are available in Safetensors format.
83
 
84
  [Download](/strangerzonehf/Flux-Icon-Kit-LoRA/tree/main) them in the Files & versions tab.
85
+
86
+ ---