Kaggle Springleaf Marketing Response Challenge
Code for 75th place (out of 2,225) in Kaggle Springleaf Marketing Response Challenge (private leaderboard).
Feature engineering (not all are used in final ensemble)
- Derived the following from date columns
- Year
- Month
- Day
- Quarter
- Day of Year
- Week of Year
- Day of Week
- Week of Month
- Hour of day (for one variable)
- One-Hot Encoding for variables likely to be categorical variables
Models
- XGBoost
Software
- Python 2.7
- Python libraries:
- Pandas
- Xgboost
- numpy