kaggle Titanic Example

1. Create Environment

$ pipenv sync

2. Prepare Dataset

$ make dataset
$ ls data
gender_submission.csv  test.csv  train.csv

3. Write Configurations

• Create features by using pdpipe: configs/pdpipeline.jsonnet
• Configure SklearnWorker: configs/lgbm.jsonnet

4. Train Model and Make Predictions

$ make run
$ head submit.csv
PassengerId,Survived
892,0
893,1
894,0
895,0
896,1
897,0
898,1
899,0
900,1