Big Mart Sales Prediction using CatBoost

About

This project implements a machine learning model using CatBoost Regressor to predict sales for Big Mart items based on various features like Item MRP, Outlet details, and establishment year. The model handles data preprocessing, missing value imputation, feature engineering, and provides a user-friendly Tkinter GUI for real-time sales prediction.

---

Features

• Data cleaning and imputation (linear interpolation, KNN imputer)
• Feature engineering and selection
• Model training and evaluation with CatBoost Regressor
• Cross-validation and performance metrics (R² score, Mean Absolute Error)
• Feature importance analysis
• GUI application for easy input and prediction display

---

Sample GUI Prediction Output

Requirements

• Python 3.x
• Libraries:
  • pandas
  • numpy
  • seaborn
  • matplotlib
  • scikit-learn
  • catboost
  • joblib
  • tkinter (usually included with Python)

Install dependencies via:

pip install pandas numpy seaborn matplotlib scikit-learn catboost joblib