I used Python 2.7 for this project. Most of the libraries used were part of scikit-learn, pandas, and numpy, although not all. I had to create an environment for XGBoost's API with scikit-learn. I included the required libraries in the Juypter Notebook which contains my code. The project is written in a PDF and follows my code in terms of flow and organization.
I analyzed 1.5 years of Santander data to predict which products existing customers use purchase next month.
Datasets used the project can be downloaded on Kaggle (test_ver2.csv, train_ver2.csv): https://www.kaggle.com/c/santander-product-recommendation/data