📊 Student Performance Predictor

A linear regression model built from scratch in Python that predicts a student's Performance Index based on study habits and lifestyle factors — using only NumPy and csv, no ML libraries.

---

Dataset

Student_Performance.csv — 10,000 student records with the following features:

Feature	Description
Hours Studied	Daily study hours
Previous Scores	Scores from previous exams
Extracurricular Activities	Yes / No (encoded to 1 / 0)
Sleep Hours	Average sleep per night
Sample Question Papers Practiced	Number of practice papers
Performance Index	Target variable (0–100)

---

How to Run

Make sure you have Python 3 and NumPy installed.

pip install numpy
python3 main.py

---

How It Works

1. Loads and preprocesses the CSV (Yes/No encoded to 1/0)
2. Normalizes all features using mean and standard deviation
3. Initializes weights and bias to zero
4. Computes gradients manually using batch gradient descent
5. Updates weights and bias each epoch
6. Prints cost every 100 epochs to track training progress

---

Project Structure

main.py                     # Model implementation
Student_Performance.csv     # Dataset
README.md                   # Project documentation

---

Built With

• Python 3
• NumPy
• Inspired by Andrew Ng's Machine Learning Specialization (Course 1)

---

Author

Sanket Dalal
B.Tech IT — Delhi Technological University
GitHub | LinkedIn