Logo Lanfrica

AI2CUP/AI2CUP

Domain:

agriculture

Record type:

software
Creator:
AI2
Host:
AI-powered platform for improving coffee trade in Ethiopia. # β˜• AI2CUP - AI-Powered Ethiopian Coffee Trade Platform > **MVP / Demo-Ready** - Predict fair prices, verify bean quality, and connect with trading partners using artificial intelligence. --- ## πŸ“‹ Features | Feature | Description | |---------|-------------| | **πŸ“ˆ Price Prediction** | ML-powered coffee price prediction based on region, month, altitude, rainfall, and variety | | **πŸ”¬ Quality Detection** | Upload a coffee bean image and get instant quality grade (High / Medium / Low) | | **πŸͺ Marketplace** | Browse sellers and buyers with smart matching | --- ## πŸ—‚ Project Structure ``` AI2CUP/ β”œβ”€β”€ backend/ β”‚ β”œβ”€β”€ main.py # FastAPI entry point β”‚ β”œβ”€β”€ requirements.txt # Python dependencies β”‚ β”œβ”€β”€ data/ β”‚ β”‚ β”œβ”€β”€ generate_dataset.py # Synthetic dataset generator β”‚ β”‚ └── coffee_prices.csv # Generated dataset (auto-created) β”‚ β”œβ”€β”€ ml/ β”‚ β”‚ β”œβ”€β”€ price_model.py # Linear Regression price model β”‚ β”‚ β”œβ”€β”€ quality_detector.py # Image quality analysis (simulated) β”‚ β”‚ β”œβ”€β”€ matcher.py # Buyer/Seller matching engine β”‚ β”‚ └── trained_model.joblib # Trained model (auto-created) β”‚ └── routes/ β”‚ β”œβ”€β”€ price.py # /predict-price endpoint β”‚ β”œβ”€β”€ quality.py # /analyze-quality endpoint β”‚ └── match.py # /match endpoints β”œβ”€β”€ frontend/ β”‚ β”œβ”€β”€ index.html # Main UI β”‚ β”œβ”€β”€ script.js # Frontend logic & API calls β”‚ └── styles.css # Premium dark theme └── README.md ``` --- ## πŸš€ Quick Start ### Prerequisites - **Python 3.9+** installed - **pip** package manager ### 1. Install Dependencies ```bash cd backend pip install -r requirements.txt ``` ### 2. Run the Server ```bash cd backend uvicorn main:app --reload --port 8000 ``` The server will automatically: - βœ… Generate the synthetic coffee price dataset - βœ… Train the ML model - βœ… Start serving the API and frontend ### 3. Open the App Visit **http://localhost:8000** in your b …

Languages