| 65 | |
| 66 | # This loop essentially from Paul's starter code |
| 67 | def cv_loop(X, y, model, N): |
| 68 | mean_auc = 0. |
| 69 | for i in range(N): |
| 70 | X_train, X_cv, y_train, y_cv = cross_validation.train_test_split( |
| 71 | X, y, test_size=.20, |
| 72 | random_state = i*SEED) |
| 73 | model.fit(X_train, y_train) |
| 74 | preds = model.predict_proba(X_cv)[:,1] |
| 75 | auc = metrics.auc_score(y_cv, preds) |
| 76 | print "AUC (fold %d/%d): %f" % (i + 1, N, auc) |
| 77 | mean_auc += auc |
| 78 | return mean_auc/N |
| 79 | |
| 80 | def main(train='train.csv', test='test.csv', submit='logistic_pred.csv'): |
| 81 | print "Reading dataset..." |