Построение модели с использованием полиномиальной функции

from sklearn import datasets, linear_model
from sklearn.metrics import r2_score, mean_squared_error
from sklearn.preprocessing import PolynomialFeatures
from sklearn.linear_model import LinearRegression
from sklearn.pipeline import Pipeline
from sklearn.model_selection import cross_val_score
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
 
## подкл выборку
 
x_train = pd.read_table ("tic/ticdata2000.txt", header = None).iloc [0:4000, 0:85]
x_test = x_train.iloc[0:4000, 84]
 
 
y_train = pd.read_table ("tic/ticeval2000.txt", header = None)
y_test = pd.read_table ("tic/tictgts2000.txt", header = None)
 
 
## обучение выборки
regression = linear_model.LinearRegression()
regression.fit (x_train, y_train)
 
 
train_predict = regression.predict (x_train)
 
print('Коэффициенты: \n', regression.coef_)
print("Cреднеквадратичная ошибка: %.2f" % mean_squared_error(y_train, train_predict))
print('Оценка отклонения: %.2f' % r2_score(y_train, train_predict))
 
## Проверка точности модели по тестовой выборке и запись в результирующий файл
y_pr = pd.DataFrame(train_predict)
y_test = y_test.reset_index(drop = True)
res = pd.concat([y_pr, y_test], axis=1)
res.to_csv("result.txt", index = False)
 
degrees = [1,5]
err = [1,5]
x_train_pol = x_train
y_train_pol = x_test
x_test_pol = y_train
y_test_pol = y_test
 
for i in range(len(degrees)):
    polynomial_features = PolynomialFeatures(degree = degrees[i],
                                             include_bias = False)
    linear_regression = LinearRegression()
    pipeline = Pipeline([("polynomial_features", polynomial_features),
                         ("linear_regression", linear_regression)])
    pipeline.fit(x_train_pol, y_train_pol)
    scores = cross_val_score(pipeline, x_train_pol, y_train_pol,
                             scoring="neg_mean_squared_error")
    pred_y = pipeline.predict(x_test_pol)
    err[i] = -scores.mean()
    print("\nСтепень: {}\nСреднеквадратичная ошибка = {}(+/- {})".format(degrees[i], -scores.mean(),
                                               scores.std()))
    print('Показатель отклонения: %.3f' % r2_score(y_test_pol, pred_y))

MemoryError                               Traceback (most recent call last)
<ipython-input-5-84ddafffa438> in <module>()
     12     pipeline = Pipeline([("polynomial_features", polynomial_features),
     13                          ("linear_regression", linear_regression)])
---> 14     pipeline.fit(x_train_pol, y_train_pol)
     15     scores = cross_val_score(pipeline, x_train_pol, y_train_pol,
     16                              scoring="neg_mean_squared_error")
 
/usr/local/lib/python3.5/dist-packages/sklearn/pipeline.py in fit(self, X, y, **fit_params)
    255             This estimator
    256         """
--> 257         Xt, fit_params = self._fit(X, y, **fit_params)
    258         if self._final_estimator is not None:
    259             self._final_estimator.fit(Xt, y, **fit_params)
 
/usr/local/lib/python3.5/dist-packages/sklearn/pipeline.py in _fit(self, X, y, **fit_params)
    220                 Xt, fitted_transformer = fit_transform_one_cached(
    221                     cloned_transformer, None, Xt, y,
--> 222                     **fit_params_steps[name])
    223                 # Replace the transformer of the step with the fitted
    224                 # transformer. This is necessary when loading the transformer
 
/usr/local/lib/python3.5/dist-packages/sklearn/externals/joblib/memory.py in __call__(self, *args, **kwargs)
    360 
    361     def __call__(self, *args, **kwargs):
--> 362         return self.func(*args, **kwargs)
    363 
    364     def call_and_shelve(self, *args, **kwargs):
 
/usr/local/lib/python3.5/dist-packages/sklearn/pipeline.py in _fit_transform_one(transformer, weight, X, y, **fit_params)
    587                        **fit_params):
    588     if hasattr(transformer, 'fit_transform'):
--> 589         res = transformer.fit_transform(X, y, **fit_params)
    590     else:
    591         res = transformer.fit(X, y, **fit_params).transform(X)
 
/usr/local/lib/python3.5/dist-packages/sklearn/base.py in fit_transform(self, X, y, **fit_params)
    519         else:
    520             # fit method of arity 2 (supervised transformation)
--> 521             return self.fit(X, y, **fit_params).transform(X)
    522 
    523 
 
/usr/local/lib/python3.5/dist-packages/sklearn/preprocessing/data.py in transform(self, X)
   1338 
   1339         # allocate output data
-> 1340         XP = np.empty((n_samples, self.n_output_features_), dtype=X.dtype)
   1341 
   1342         combinations = self._combinations(n_features, self.degree,
 
MemoryError: