This repository contains the core components of the project focused on predicting loan defaulters using historical lending data.
Notebook/— Main analysis and modeling workflowdocs/— Project documentation outlining methodology, assumptions, and decisionsslides/— Presentation materials summarizing key findingsApp/— Contains the training data and model specifications.
- Python (Pandas, NumPy, Scikit-learn, Seaborn)
- Google Colab
- Streamlit
- End-to-end data science workflow
- Modular and reproducible code
- Deployed solution demonstrating practical application
A live version of the project is available here.
This repository is intended to showcase structure, methodology, and deployment practices. For full details, refer to the documentation and notebook.