import matplotlib.pyplot as plt
import numpy as np
from sklearn import datasets, linear_model
from sklearn.metrics import mean_squared_error
diabetes = datasets.load_diabetes()
diabetes_X = diabetes.data[:, np.newaxis,2]
diabetes_X_train = diabetes_X[:-30]
diabetes_X_test = diabetes_X[-30:]

diabetes_y_train = diabetes.target[:-30]
diabetes_y_test = diabetes.target[-30:]
model = linear_model.LinearRegression()
model.fit(diabetes_X_train,diabetes_y_train)
diabetes_y_predicted = model.predict(diabetes_X_test)
#(['data', 'target', 'frame', 'DESCR', 'feature_names', 'data_filename', 'target_filename', 'data_module'])

print("mean squared error is:", mean_squared_error(diabetes_y_test,diabetes_y_predicted))
print("weight:",model.coef_)
print("Intercept:", model.intercept_)
plt.scatter(diabetes_X_test,diabetes_y_test)
plt.plot(diabetes_X_test,diabetes_y_predicted)
plt.show()