Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
datasets:
|
4 |
+
- fhai50032/ControlNet-Poster
|
5 |
+
base_model:
|
6 |
+
- black-forest-labs/FLUX.1-dev
|
7 |
+
---
|
8 |
+
|
9 |
+
|
10 |
+
# Flux-ControlNet: Text-to-Image Diffusion Model with Caption Alignment
|
11 |
+
|
12 |
+
This repository hosts **Flux-ControlNet**, a customized ControlNet-based diffusion model designed for generating text-embedded images.
|
13 |
+
|
14 |
+
---
|
15 |
+
|
16 |
+
## Key Features
|
17 |
+
- **Flux-ControlNet**: Enhanced ControlNet architecture for better control over text-to-image generation.
|
18 |
+
- **Optimized Diffusion**: Uses Hugging Face Diffusers and Accelerate for streamlined performance.
|
19 |
+
|
20 |
+
---
|
21 |
+
|
22 |
+
## How It Works
|
23 |
+
1. **Input**: Provide text prompts and conditioning image.
|
24 |
+
2. **Processing**:
|
25 |
+
- Flux-ControlNet processes the text and applies diffusion to synthesize aligned images.
|
26 |
+
3. **Output**: High-quality, text-embedded images.
|
27 |
+
|
28 |
+
---
|
29 |
+
|
30 |
+
# Training Parameters for Flux-ControlNet
|
31 |
+
```
|
32 |
+
General Parameters:
|
33 |
+
Model Architecture: Flux-based ControlNet Model
|
34 |
+
Image Resolution: 512x512
|
35 |
+
Batch Size: 4
|
36 |
+
Epochs: 50
|
37 |
+
Optimizer: AdamW
|
38 |
+
Learning Rate: 1e-5 (with cosine schedular)
|
39 |
+
Weight Decay: 0.01
|
40 |
+
Gradient Clipping: 1.0
|
41 |
+
```
|
42 |
+
|
43 |
+
# Inference Code
|
44 |
+
Soon to be added
|