-
install python and dependencies, i.e. scikit-learn
-
git clone https://github.com/jayflo/kaggle.git
-
add
/mlcmn
toPYTHONPATH
environment variable -
install competition data to
/data
directory. See how data is loaded in each competition'smain.py
file to determine exact/data
structure, e.g.pd.read_csv('./data/train.csv'), pd.read_csv('./data/test.csv')
in
/titanic/main.py
means that competition expects the training data (from kaggle) to reside in/titanic/data/train.csv
(and similarly for the test data).