DataJam-Nov-2025

Overview

This repository contains the code and assets used for our DataJam November 2025 project.
The project focuses on training and evaluating multiple machine learning models for a classification task, including Logistic Regression, Random Forest, Gradient Boosting, XGBoost, CatBoost, and LightGBM.
The final model we used in our presentation was LightGBM with RandomizedSearchCV, which achieved the best overall performance.

Repository Structure

DataJam-Nov-2025/
│
├── Dataset/ # Dataset files used for model training/testing
├── catboost_info/ # Auto-generated CatBoost metadata folder
│
├── Imports.py # Centralized imports + helper utilities
├── main.py # Main script that loads data, preprocesses, trains models, evaluates results
│
├── requirements.txt # Python dependencies for reproducing the environment
└── .gitignore # Ignored files for version control

Installation & Setup

1. Clone the repository

git clone https://github.com/Abyss999/DataJam-Nov-2025.git
cd DataJam-Nov-2025

2. Virtual Environment

python3 -m venv datajam

Mac/Linux

source datajam/bin/activate

Windows

datajam\Scripts\activate

3. Install

pip install -r requirements.txt

4. Run Project

python main.py

This script:

• loads and cleans the dataset
• encodes categorical features
• normalizes numerical features
• trains multiple ML models
• evaluates them using accuracy, classification reports, and confusion matrices
• compares performance across models

Models Implemented

The following algorithms are implemented in main.py:

• Logistic Regression
• Decision Tree
• Random Forest
• Gradient Boosting
• XGBoost
• CatBoost
• LightGBM (best performing model with RandomizedSearchCV)