Student Exam Performance Prediction

This project is an end-to-end Machine Learning web application that predicts student exam performance based on various demographic and educational factors.

Project Overview

The goal is to understand how a student's performance (specifically in Mathematics) is affected by variables such as Gender, Ethnicity, Parental Level of Education, Lunch, and Test Preparation Course. The project implements a complete machine learning pipeline from data ingestion to deployment as a web application.

Project Flow & Pipeline

The project follows a modular architecture:

1. Data Ingestion (src/components/data_ingestion.py):

   • Reads the raw data from the source.
   • Splits the data into training and testing sets.
   • Saves the raw, train, and test data paths.

2. Data Transformation (src/components/data_transformation.py):

   • Handles missing values.
   • Performs one-hot encoding for categorical variables.
   • Standardizes numerical features using StandardScaler.
   • Saves the preprocessor object as a pickle file.

3. Model Training (src/components/model_trainer.py):

   • Trains multiple machine learning models.
   • Evaluates them based on R2 score.
   • Selects the best performing model (e.g., Linear Regression, Decision Tree, etc.).
   • Saves the best model as a pickle file.

4. Prediction Pipeline (src/pipeline/predict_pipeline.py):

   • CustomData class: Maps HTML form input to a DataFrame.
   • PredictPipeline class: Loads the saved model and preprocessor to predict the target variable.

5. Web Application (app.py):

   • A Flask web server that handles user requests.
   • Renders the UI (templates/).
   • Takes user input and calls the prediction pipeline.

Folder Structure

mlproject/
├── .ebextensions/       # AWS Elastic Beanstalk configuration
├── artifacts/           # Stores models and data (train.csv, test.csv, model.pkl, etc.)
├── catboost_info/       # CatBoost model logs
├── logs/                # Application logs
├── notebook/            # Jupyter notebooks for EDA and experimentation
├── src/                 # Source code
│   ├── components/      # ML pipeline components
│   │   ├── data_ingestion.py
│   │   ├── data_transformation.py
│   │   └── model_trainer.py
│   ├── pipeline/        # Training and prediction pipelines
│   │   ├── predict_pipeline.py
│   │   └── train_pipeline.py
│   ├── utils/           # Utility functions (save/load object, etc.)
│   ├── exception.py     # Custom exception handling
│   └── logger.py        # Logging configuration
├── static/              # Static files (CSS, JS, Images)
├── templates/           # HTML templates
│   ├── home.html        # Prediction page
│   └── index.html       # Landing page
├── app.py               # Flask application entry point
├── application.py       # WSGI entry point
├── requirements.txt     # Python dependencies
├── setup.py             # Package installation script
└── README.md            # Project documentation

How to Run

1. Install Dependencies:

   pip install -r requirements.txt

2. Run the Application:

   python app.py

3. Access the App: Open your browser and navigate to http://127.0.0.1:5000