Agricultural Price Forecasting System (AgriPrice Predict)

AgriPrice Predict is an enterprise-grade forecasting platform designed to mitigate supply chain risks in the agricultural sector. It combines a robust Python-based predictive engine with a modern Next.js dashboard to provide real-time price intelligence and actionable business strategies.

🏗️ Architecture Overview

• Backend: FastAPI (Python 3.12) serving a machine learning pipeline that includes SARIMA, XGBoost, and LSTM models.
• Frontend: Next.js (TypeScript) dashboard utilizing Recharts for visualization and Tailwind CSS for a premium UI.
• Data Pipeline: Automated cleaning, EDA, model training, and evaluation cycles.
• Deployment: Fully containerized with Docker and Docker Compose.

🚀 Getting Started

Prerequisites

• Docker & Docker Compose
• Python 3.12+ (if running locally without Docker)
• Node.js 20+ (if running locally without Docker)

Quick Start (Docker)

The easiest way to run the entire stack is using Docker Compose:

docker-compose up --build

• Frontend: http://localhost:3000
• Backend API: http://localhost:8000
• API Docs: http://localhost:8000/docs

Local Development Setup

Backend

1. Navigate to the backend/ directory.
2. Install dependencies using uv or pip:

   uv sync
   # OR
   pip install -r requirements.txt

3. Run the data pipeline to generate models and forecasts:

   python main.py pipeline

4. Start the FastAPI server:

   python main.py server

Frontend

1. Navigate to the frontend/ directory.
2. Install dependencies:

   npm install

3. Start the development server:

   npm run dev

📊 Data Pipeline & Models

The system follows a sequential pipeline:

1. Cleaning (Q1): Handles missing values via time-interpolation and removes outliers using the IQR method.
2. EDA & Training (Q2): Performs seasonal decomposition and trains three models:
   • SARIMA: Statistical model for capturing seasonality.
   • XGBoost: Gradient boosted trees for capturing non-linear patterns.
   • LSTM: Deep learning model for long-term sequence dependencies.
3. Evaluation (Q3): Compares models using MAE, RMSE, and MAPE metrics.
4. Integration (Q7): Exposes the XGBoost model via a REST API for real-time inference.

🛠️ Production Readiness

Environment Variables

Configure these in your production environment or .env files:

Variable	Description	Default
PORT	Backend server port	8000
ALLOWED_ORIGINS	CORS allowed origins (comma-separated)	*
DATA_RAW	Path to raw input CSV	data/raw/agricultural_prices.csv
MODELS_DIR	Directory to store trained models	models/
NEXT_PUBLIC_API_URL	Frontend pointer to Backend API	http://localhost:8000

Security

• CORS: In production, set ALLOWED_ORIGINS to your frontend domain.
• Secrets: Change SECRET_KEY in backend/.env to a secure random string.
• Docker: The provided Dockerfiles use multi-stage builds and non-root users (where applicable) for security.

Monitoring

• Health Check: Monitor GET /health for backend status.
• Logs: Backend uses standard Python logging for easier integration with ELK or CloudWatch.

📈 Dashboard Features

• Price Trends: Interactive line charts showing historical vs. forecasted prices.
• What-if Analysis: A custom prediction form to test "what-if" scenarios based on the last 12 months of price data.
• Risk Mitigation: AI-driven strategy cards providing procurement and supply chain advice based on volatility forecasts.

🌐 Deployment

Backend (Render.com)

This project is pre-configured for Render.com using the render.yaml blueprint.

1. Connect your GitHub repository to Render.
2. Select "Blueprint" and choose Agriculture-Forecasting.
3. Render will automatically build the Docker container and deploy it.
4. Important: The free tier has 512MB RAM. The Dockerfile is optimized to use tensorflow-cpu and a single Gunicorn worker to stay within these limits.

Frontend (Vercel)

The Next.js frontend is best deployed on Vercel.

1. Connect your GitHub repository to Vercel.
2. Add an Environment Variable:
   • NEXT_PUBLIC_API_URL: Your Render backend URL (e.g., https://agri-forecast-backend.onrender.com).
3. Vercel will handle the build and deployment automatically.

🤝 Contributing

1. Fork the repo.
2. Create your feature branch (git checkout -b feature/AmazingFeature).
3. Commit your changes (git commit -m 'Add some AmazingFeature').
4. Push to the branch (git push origin feature/AmazingFeature).
5. Open a Pull Request.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.