Calories Burnt Prediction Using Flask and Machine Learning

This project is a web-based application that predicts the calories burnt based on user inputs using a machine learning model trained with XGBRegressor. The application is built with Python, Flask, and scikit-learn, providing an interactive and user-friendly experience.

Overview

The Calories Burnt Prediction application enables users to calculate their calorie expenditure by inputting details such as age, gender, weight, height, and activity level. The backend uses a machine learning model trained on real-world data to make accurate predictions.

This project demonstrates how machine learning models can be deployed in a lightweight web application using Flask, making it ideal for learning and practical usage.

---

Features

• Accurate Predictions: Utilizes XGBRegressor for reliable calorie prediction.
• Interactive User Interface: HTML and CSS ensure a responsive and visually appealing design.
• Machine Learning Backend: Integrates a pre-trained model (calories_model.pkl) for predictions.
• Flask Framework: Provides a lightweight and fast backend framework.
• Scalable Deployment: Configured for deployment on Render using Gunicorn.

---

Machine Learning Model

• Algorithm: XGBRegressor (Extreme Gradient Boosting)
• Data Preprocessing: Used techniques like normalization and imputation for clean and efficient data handling.
• Training: The model was trained using scikit-learn and XGBoost to ensure high accuracy.
• Saved Model: The trained model is saved in calories_model.pkl for quick loading and inference.

---