🎓 Student Exam Performance Predictor

An end-to-end Machine Learning web application that predicts a student’s Maths score based on demographic information, academic background, and preparation level.
The project demonstrates how a trained ML model can be deployed as a production-ready Flask web application.

---

🚀 Live Demo

🔗 Deployed Application:
https://student-performance-indicator-0w7h.onrender.com

---

📌 Project Overview

This project is designed to showcase:

• Complete Machine Learning lifecycle
• Modular coding practices
• Model serialization and reuse
• Web-based inference using Flask
• Clean and interactive UI

The application takes user inputs through a web form and returns a real-time prediction using a trained regression model.

---

🧠 Problem Statement

Given a student’s:

• Gender
• Race / Ethnicity
• Parental level of education
• Lunch type
• Test preparation status
• Reading score
• Writing score

Predict the Maths score using a Machine Learning model.

---

🏗️ Project Architecture

├── ./ │ ├── app.py │ ├── README.md │ ├── requirements.txt │ ├── setup.py │ ├── tree.py │ ├── logs/ │ │ ├── 01_20_2026_22_47_14.log │ ├── ML_Project.egg-info/ │ │ ├── dependency_links.txt │ │ ├── PKG-INFO │ │ ├── requires.txt │ │ ├── SOURCES.txt │ │ ├── top_level.txt │ ├── Notebook/ │ │ ├── EDA.ipynb │ │ ├── model_training.ipynb │ │ ├── data/ │ │ │ ├── stud.csv │ ├── src/ │ │ ├── exception.py │ │ ├── logger.py │ │ ├── utils.py │ │ ├── init.py │ │ ├── components/ │ │ │ ├── data_ingestion.py │ │ │ ├── data_transformation.py │ │ │ ├── model_trainer.py │ │ │ ├── init.py │ │ ├── pipeline/ │ │ │ ├── predict_pipeline.py │ │ │ ├── train_pipeline.py │ │ │ ├── init.py │ ├── templates/ │ │ ├── home.html │ │ ├── index.html

---

⚙️ Tech Stack

🔹 Programming & Libraries

• Python
• NumPy
• Pandas
• Scikit-learn

🔹 Web Framework

• Flask
• Gunicorn

🔹 Machine Learning

• Linear Regression
• Data Preprocessing Pipelines
• OneHotEncoder
• StandardScaler
• ColumnTransformer

🔹 Deployment

• Render (Cloud Hosting)

---

🧩 Machine Learning Pipeline

1. Data Ingestion
   • Raw dataset loading
2. Data Transformation
   • Categorical encoding
   • Numerical scaling
3. Model Training
   • Multiple regression models tested
   • Best model selected using R² score
4. Serialization
   • Model and preprocessor saved using Pickle
5. Inference
   • Same preprocessor reused during prediction to ensure consistency

---

🖥️ Web Application Flow

1. User lands on a Landing Page (/)
2. Redirects to the Prediction Page (/predictdata)
3. User fills the form
4. Flask backend:
   • Converts inputs to DataFrame
   • Applies preprocessing
   • Runs model inference
5. Predicted Maths score is displayed on the UI