The code was written for Loan Default Prediction Competition at Kaggle and got the prize.
pandas: version 0.13.1 or later
scikit learn: dev branch with version commit 884889a4cd36e63d53a067d9380dea7724a93ac5 or later
- Download data from Kaggle
- Unzip the train and test csv files to path/to/data/folder and make sure that their names are train_v2.csv and test_v2.csv, respectively
- Run
python train_predict.py path/to/data/folder
- The prediction submission-ready csv (submission.csv) will be found at path/to/data/folder