🔥 Calories Burned Prediction using Machine Learning

📘 Project Overview

This project predicts the number of calories burned during physical exercise using Machine Learning.
It leverages physiological and workout-related parameters such as Gender, Age, Height, Weight, Duration, Heart Rate, and Body Temperature to generate accurate, data-driven predictions.

The system is implemented using Python and Streamlit, and deployed as a web application that provides real-time calorie predictions with visual insights.

---

🧠 Key Features

• ✅ Predicts calories burned based on user inputs.
• ✅ Uses Random Forest Regressor (best model) for high accuracy.
• ✅ Displays feature importance and prediction visualizations.
• ✅ Interactive Streamlit web interface.
• ✅ Easy deployment — works locally or via Streamlit Cloud.

---

⚙️ Workflow

1. Data Collection – Combined exercise.csv and calories.csv datasets.
2. Preprocessing – Encoding gender, scaling numerical values, and removing non-predictive columns.
3. Model Training – Trained Linear Regression and Random Forest models.
4. Evaluation – Compared models using MAE, MSE, RMSE, and R² Score.
5. Deployment – Built a Streamlit app for real-time prediction.

---

🧩 Technologies Used

Component	Tool / Library
Language	Python 3
Web Framework	Streamlit
ML Library	Scikit-learn
Data Handling	Pandas, NumPy
Visualization	Matplotlib, Seaborn

---

📊 Model Performance

Model	MAE	MSE	RMSE	R² Score
Linear Regression	8.44	131.99	11.49	0.967
Random Forest Regressor	1.71	7.13	2.67	0.998

✅ Random Forest was selected as the final model due to superior accuracy and generalization.

---

🚀 How to Run Locally

1️⃣ Clone the repository

git clone https://github.com/PKrishnaS/calories-burner.git
cd calories-burner