if __name__ == '__main__':
X, y = OttoCompetition.load_data(train=True)
X_test, _ = OttoCompetition.load_data(train=False)
le = LabelEncoder().fit(y)
all_hold = []
all_hold_predict = []
all_test_predict = []
all_weights = []
# 20% holdout
for i, (data_index, hold_index) in enumerate(StratifiedKFold(y, n_folds = 5, random_state=0)):
X_data, X_hold = X[data_index], X[hold_index]
y_data, y_hold = y[data_index], y[hold_index]
y_hold_predict = []
y_test_predict = []
# train with 50%, validation with 5%
for j, (train_index, valid_index) in enumerate(StratifiedShuffleSplit(y_data, 20, test_size = 0.05, train_size = 0.5, random_state=0)):
X_train, X_valid = X_data[train_index], X_data[valid_index]
y_train, y_valid = y_data[train_index], y_data[valid_index]
clf, evals = train_xgboost(X_train, le.transform(y_train),
X_valid, le.transform(y_valid))
y_hold_predict.append(clf.predict(xgb.DMatrix(X_hold)))
y_test_predict.append(clf.predict(xgb.DMatrix(X_test)))
weights = OttoCompetition.prediction_weights(y_hold_predict, y_hold)
all_hold.append(y_hold)
all_hold_predict.append(y_hold_predict)
