# predict scores
y_treek_train_pred = treek.predict(x_traink)
y_treek_test_pred = treek.predict(x_testk)
# Performance scores (for train, test dataset r2 scores and mean squared error)
print('\nkarar ağaçları yeni özniteliklerle:\n')
print('Eğitim performansı (R kare): ', r2_score(y_traink, y_treek_train_pred))
print('Test performansı (R kare): ', r2_score(y_testk, y_treek_test_pred))
print('Eğitim hata oranı (MSE): ', mean_squared_error(y_traink, y_treek_train_pred))
print('Test hata oranı (MSE): ', mean_squared_error(y_testk, y_treek_test_pred))
# Parameter optimization process
# we doing paramater optimazation process for the all machine learning models
# Polinomial Regresion model Parameter optimization
print('\nparametre optimizasyonu\n')
print('Polinom Modelinin parametreleri: \n', polyk._get_param_names())
import time
print('Polinom Dereceleri\n')
for a in range(1,5):
tic=time.time()
polinom_opt_pred = PolynomialFeatures(degree=a)
x_poly_opt = polinom_opt_pred.fit_transform(x_kbest)
x_train1, x_test1, y_train1, y_test1 = train_test_split(x_poly_opt, y, test_size =0.3, random_state=0)
polimodelopt = LinearRegression()
polimodelopt.fit(x_train1, y_train1)
pred_poly_y_train1= polimodelopt.predict(x_train1)
pred_poly_y_test1= polimodelopt.predict(x_test1)
print('derece ', a, 'için :\n')
print('MSE Eğitim: ', mean_squared_error(y_train1, pred_poly_y_train1))
print('MSE Test: ', mean_squared_error(y_test1, pred_poly_y_test1))
print('R Kare Eğitim Skoru: ', r2_score(y_train1, pred_poly_y_train1))
