Upload QUICK_START.md with huggingface_hub
Browse files- QUICK_START.md +81 -0
QUICK_START.md
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# TRIQA Hugging Face Demo - Quick Start Guide
|
| 2 |
+
|
| 3 |
+
## 🚀 Create Your Hugging Face Space
|
| 4 |
+
|
| 5 |
+
### Step 1: Create the Space
|
| 6 |
+
1. Go to https://huggingface.co/spaces
|
| 7 |
+
2. Click "Create new Space"
|
| 8 |
+
3. Fill in:
|
| 9 |
+
- **Space name**: `triqa-image-quality-assessment`
|
| 10 |
+
- **License**: MIT
|
| 11 |
+
- **SDK**: Gradio
|
| 12 |
+
- **Hardware**: CPU (or GPU if available)
|
| 13 |
+
- **Visibility**: Public
|
| 14 |
+
|
| 15 |
+
### Step 2: Upload Files
|
| 16 |
+
Upload these files from this folder to your Hugging Face Space:
|
| 17 |
+
|
| 18 |
+
#### Core Files (Upload these first):
|
| 19 |
+
- `app.py` - Main application
|
| 20 |
+
- `README.md` - Space description
|
| 21 |
+
- `requirements.txt` - Dependencies
|
| 22 |
+
- `package.json` - Node.js config
|
| 23 |
+
- `convnext_original.py` - Model architecture
|
| 24 |
+
- `convnext_finetune.py` - Model architecture
|
| 25 |
+
- `sample_image/` folder - Sample images
|
| 26 |
+
|
| 27 |
+
#### Model Files (Download from Box first):
|
| 28 |
+
1. Go to: https://utexas.box.com/s/8aw6axc2lofouja65uc726lca8b1cduf
|
| 29 |
+
2. Download the model files
|
| 30 |
+
3. Create these directories in your Space:
|
| 31 |
+
- `feature_models/`
|
| 32 |
+
- `Regression_Models/`
|
| 33 |
+
4. Upload the model files to their respective directories
|
| 34 |
+
|
| 35 |
+
### Step 3: Test the Demo
|
| 36 |
+
1. Wait for the build to complete (2-5 minutes)
|
| 37 |
+
2. Upload a sample image
|
| 38 |
+
3. Click "Assess Quality"
|
| 39 |
+
4. Verify you get a quality score (1-5 scale)
|
| 40 |
+
|
| 41 |
+
## 📁 File Structure in Your Space
|
| 42 |
+
```
|
| 43 |
+
your-space/
|
| 44 |
+
├── app.py
|
| 45 |
+
├── README.md
|
| 46 |
+
├── requirements.txt
|
| 47 |
+
├── package.json
|
| 48 |
+
├── convnext_original.py
|
| 49 |
+
├── convnext_finetune.py
|
| 50 |
+
├── sample_image/
|
| 51 |
+
│ ├── 233045618.jpg
|
| 52 |
+
│ ├── 25239707.jpg
|
| 53 |
+
│ ├── 44009500.jpg
|
| 54 |
+
│ ├── 5129172.jpg
|
| 55 |
+
│ └── 85119046.jpg
|
| 56 |
+
├── feature_models/
|
| 57 |
+
│ ├── convnext_tiny_22k_224.pth
|
| 58 |
+
│ └── triqa_quality_aware.pth
|
| 59 |
+
└── Regression_Models/
|
| 60 |
+
├── KonIQ_scaler.save
|
| 61 |
+
└── KonIQ_TRIQA.save
|
| 62 |
+
```
|
| 63 |
+
|
| 64 |
+
## 🎯 Expected Demo Features
|
| 65 |
+
- **Image Upload**: Drag & drop or click to upload
|
| 66 |
+
- **Quality Assessment**: Click button to get score
|
| 67 |
+
- **Sample Images**: Pre-loaded examples
|
| 68 |
+
- **Paper Links**: arXiv and IEEE links
|
| 69 |
+
- **Citation**: BibTeX citation
|
| 70 |
+
|
| 71 |
+
## 🔧 Troubleshooting
|
| 72 |
+
- **Build fails**: Check requirements.txt
|
| 73 |
+
- **Models not loading**: Verify model files are uploaded
|
| 74 |
+
- **File not found**: Check file paths and names
|
| 75 |
+
- **Memory issues**: Use CPU hardware
|
| 76 |
+
|
| 77 |
+
## 📞 Support
|
| 78 |
+
- GitHub: https://github.com/rajeshsureddi/triqa
|
| 79 |
+
- Paper: https://arxiv.org/pdf/2507.12687
|
| 80 |
+
|
| 81 |
+
Your demo will be live at: `https://huggingface.co/spaces/your-username/triqa-image-quality-assessment`
|