mse = mean_squared_error(y_test, est.predict(X_test))
print('mean squared error = {}'.format(mse))
y_pred = est.predict(X_test)
r_squared = est.score(X_test, y_test)
print('r squared = {}'.format(r_squared))
else:
mse = mean_squared_error(y_test, est.predict(X_test_std))
print('mean squared error = {}'.format(mse))
y_pred = est.predict(X_test_std)
r_squared = est.score(X_test_std, y_test)
print('r squared = {}'.format(r_squared))
#
test_score = np.zeros((n,), dtype=np.float64)
for i, y_pred in enumerate(est.staged_predict(X_test)):
test_score[i] = est.loss_(y_test, y_pred)
plt.figure(figsize=(12, 6))
plt.subplot(1, 2, 1)
plt.title('Deviance')
plt.plot(np.arange(n) + 1, est.train_score_, 'b-',
label='Training Set Deviance')
plt.plot(np.arange(n) + 1, test_score, 'r-',
label='Test Set Deviance')
plt.legend(loc='upper right')
plt.xlabel('Boosting Iterations')
plt.ylabel('Deviance')
# #############################################################################
# Plot feature importance
feature_importance = est.feature_importances_
