Ever wondered "What breed is that adorable dog?" π€
DoggyDex is your AI-powered dog breed identification companion that can instantly recognize over 100+ dog breeds from a single photo! Built with cutting-edge machine learning and a sleek web interface, it's like having a canine expert in your pocket.
|
|
# Clone the repository
git clone https://github.com/yourusername/doggydex.git
cd doggydex
# Build and run with Docker
docker build -t doggydex .
docker run -p 5001:5001 doggydex# Install dependencies with Poetry
poetry install
# Activate virtual environment
poetry shell
# Run the application
python app.pyVisit http://localhost:5001 and start identifying dogs! π
| Component | Version | Purpose |
|---|---|---|
| π Python | 3.10+ | Core runtime |
| π§ PyTorch | Latest | ML inference |
| π Flask | 3.1+ | Web framework |
| π§ Poetry | Latest | Dependency management |
doggydex/
βββ π app.py # Application entry point
βββ π§ pyproject.toml # Project configuration
βββ π³ Dockerfile # Container configuration
βββ π§ models/ # ML model files
β βββ dognet-convnext_large.pth
β βββ dognext_large_quantized.pth
βββ π website/ # Web application
β βββ π¨ static/ # CSS, JS, images
β βββ π templates/ # HTML templates
β βββ π identifier.py # ML inference logic
β βββ π― views.py # Route handlers
βββ π certs/ # SSL certificates
Clean, modern interface ready for your dog photos
Instant breed identification with confidence scores
Fully responsive design for on-the-go identification
graph LR
A[πΈ Upload Photo] --> B[π Preprocess Image]
B --> C[π§ ConvNeXt Model]
C --> D[π Confidence Score]
D --> E[β¨ Display Results]
- πΈ Upload - Take or select your dog photo
- π Process - Image preprocessing and normalization
- π§ Analyze - ConvNeXt Large model inference
- π Score - Confidence percentage calculation
- β¨ Results - Beautiful breed identification display
- π― Accurate: Trained on thousands of dog images
- β‘ Fast: Optimized for real-time inference
- π¨ Beautiful: Modern, responsive design
- π Secure: HTTPS-ready with SSL support
- π³ Scalable: Docker containerization
- π Free: Open source and free to use
We love contributions! π
- π Report bugs - Help us improve
- β¨ Suggest features - Share your ideas
- π Improve docs - Make it clearer
- π§ Train models - Better accuracy
- π¨ Design UI - Enhanced experience
This project is licensed under the MIT License - see the LICENSE file for details.