KardomAi - Cardamom Price Prediction

A machine learning project that predicts cardamom prices using time series analysis with SARIMA (Seasonal AutoRegressive Integrated Moving Average) model. The project combines web-scraped market data with advanced time series analysis to provide accurate price forecasts.

Features

• Time Series Analysis: Advanced SARIMA modeling for price prediction
• Web Scraping: Automated data collection from Indian Spices Board
• Data Visualization: Interactive plots of actual vs predicted prices
• Price Forecasting: Future price predictions with confidence intervals

Installation

1. Clone the repository:

git clone https://github.com/yourusername/kardomai.git
cd kardomai

2. Install required packages:

pip install pandas numpy matplotlib statsmodels selenium scikit-learn

Usage

Price Prediction

from tsa import KardomAi

# Initialize and train model
predictor = KardomAi("cardamom data.xlsx")
predictor.split_data(test_size=0.2)
predictor.train()

# Make predictions
results = predictor.evaluate()
future_prices = predictor.predict_future(
    start_date="2024-03-01",
    end_date="2024-03-31"
)

# Visualize results
from tsa import plot_actual_vs_predicted
plot_actual_vs_predicted(predictor, results, future_prices)

Web Scraping

from webscraping import scrape_prices

# Scrape latest prices from Indian Spices Board
scrape_prices()  # Saves data to 'scraped_data.csv'

Project Structure

• tsa.py - Core time series analysis module with SARIMA implementation
• webscraping.py - Web scraping module for Indian Spices Board data
• README.md - Project documentation
• cardamom data.xlsx - Historical price data

Data Requirements

The input Excel file should contain:

• Date of Auction - Date in YYYY-MM-DD format
• MaxPrice (Rs./Kg) - Daily maximum price in Indian Rupees per Kilogram

Model Details

The SARIMA model implementation includes:

• Automatic parameter optimization
• Stationarity testing
• Seasonal decomposition
• Cross-validation
• Performance metrics (RMSE, MAE)

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

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

Author

Raynell Rajeev

Acknowledgments

• Indian Spices Board for providing market data
• Open source community for tools and libraries