Skip to content

Latest commit

 

History

History
49 lines (30 loc) · 1.94 KB

README.md

File metadata and controls

49 lines (30 loc) · 1.94 KB
Raw

Heat Index Forecasting App

Overview

Welcome to the Heat Index Forecasting App repository! This application is designed to predict the heat index for various regions in the Philippines. Using historical weather data, this model forecasts the heat index for the current day and the next 7 days using four (4) algorithms, namely "Linear Regression", "Decision Tree", "k-NN", and "Random Forest Regression". This backend service provides predictions via a RESTful API for use in mobile applications and utilizes the algorithms for modeling and forecasting based on historical weather data from PAGASA with parameters "Year", "Month", "Day", "HeatIndex" (celsius).

Features

  • Predicts Heat Index: Forecasts the current heat index for selected regions based on historical data.
  • 7-Day Forecast: Provides heat index values for the next 7 days.
  • Region-Specific Models: Utilizes machine learning models tailored to different regions in the Philippines.

Project Structure

  • backend/: Contains the backend code and model files.
    • app.py: Main application script for serving the API.
    • Data/: Historical weather data.
    • models/: Saved machine learning models and scalers.
  • requirements.txt: Python dependencies for the project.

Installation

To set up the project locally, follow these steps:

  1. Clone the repository:

    git clone https://github.com/carolbanal/heat-index-forecasting-app.git
cd heat-index-forecasting-app

  2. Install dependencies: It is recommended to create a virtual environment before installing dependencies.

    python -m venv venv
source venv/bin/activate  # On Windows, use `venv\Scripts\activate`
pip install -r requirements.txt

  3. Set up the environment: Ensure that you have the necessary environment variables and configuration files set up. Refer to .env.example for details.

  4. Run the application:

    python backend/app.py