In [3]:

    

import sklearn


    

In [4]:

    

from sklearn import datasets


    

In [5]:

    

diabetes = datasets.load_diabetes()


    

In [7]:

    

from sklearn import linear_model


    

In [8]:

    

import numpy as np
diabetes_X = diabetes.data[:, np.newaxis, 2]


    

In [9]:

    

diabetes_X


    

    
Out[9]:





array([[ 0.06169621],
       [-0.05147406],
       [ 0.04445121],
       [-0.01159501],
       [-0.03638469],
       [-0.04069594],
       [-0.04716281],
       [-0.00189471],
       [ 0.06169621],
       [ 0.03906215],
       [-0.08380842],
       [ 0.01750591],
       [-0.02884001],
       [-0.00189471],
       [-0.02560657],
       [-0.01806189],
       [ 0.04229559],
       [ 0.01211685],
       [-0.0105172 ],
       [-0.01806189],
       [-0.05686312],
       [-0.02237314],
       [-0.00405033],
       [ 0.06061839],
       [ 0.03582872],
       [-0.01267283],
       [-0.07734155],
       [ 0.05954058],
       [-0.02129532],
       [-0.00620595],
       [ 0.04445121],
       [-0.06548562],
       [ 0.12528712],
       [-0.05039625],
       [-0.06332999],
       [-0.03099563],
       [ 0.02289497],
       [ 0.01103904],
       [ 0.07139652],
       [ 0.01427248],
       [-0.00836158],
       [-0.06764124],
       [-0.0105172 ],
       [-0.02345095],
       [ 0.06816308],
       [-0.03530688],
       [-0.01159501],
       [-0.0730303 ],
       [-0.04177375],
       [ 0.01427248],
       [-0.00728377],
       [ 0.0164281 ],
       [-0.00943939],
       [-0.01590626],
       [ 0.0250506 ],
       [-0.04931844],
       [ 0.04121778],
       [-0.06332999],
       [-0.06440781],
       [-0.02560657],
       [-0.00405033],
       [ 0.00457217],
       [-0.00728377],
       [-0.0374625 ],
       [-0.02560657],
       [-0.02452876],
       [-0.01806189],
       [-0.01482845],
       [-0.02991782],
       [-0.046085  ],
       [-0.06979687],
       [ 0.03367309],
       [-0.00405033],
       [-0.02021751],
       [ 0.00241654],
       [-0.03099563],
       [ 0.02828403],
       [-0.03638469],
       [-0.05794093],
       [-0.0374625 ],
       [ 0.01211685],
       [-0.02237314],
       [-0.03530688],
       [ 0.00996123],
       [-0.03961813],
       [ 0.07139652],
       [-0.07518593],
       [-0.00620595],
       [-0.04069594],
       [-0.04824063],
       [-0.02560657],
       [ 0.0519959 ],
       [ 0.00457217],
       [-0.06440781],
       [-0.01698407],
       [-0.05794093],
       [ 0.00996123],
       [ 0.08864151],
       [-0.00512814],
       [-0.06440781],
       [ 0.01750591],
       [-0.04500719],
       [ 0.02828403],
       [ 0.04121778],
       [ 0.06492964],
       [-0.03207344],
       [-0.07626374],
       [ 0.04984027],
       [ 0.04552903],
       [-0.00943939],
       [-0.03207344],
       [ 0.00457217],
       [ 0.02073935],
       [ 0.01427248],
       [ 0.11019775],
       [ 0.00133873],
       [ 0.05846277],
       [-0.02129532],
       [-0.0105172 ],
       [-0.04716281],
       [ 0.00457217],
       [ 0.01750591],
       [ 0.08109682],
       [ 0.0347509 ],
       [ 0.02397278],
       [-0.00836158],
       [-0.06117437],
       [-0.00189471],
       [-0.06225218],
       [ 0.0164281 ],
       [ 0.09618619],
       [-0.06979687],
       [-0.02129532],
       [-0.05362969],
       [ 0.0433734 ],
       [ 0.05630715],
       [-0.0816528 ],
       [ 0.04984027],
       [ 0.11127556],
       [ 0.06169621],
       [ 0.01427248],
       [ 0.04768465],
       [ 0.01211685],
       [ 0.00564998],
       [ 0.04660684],
       [ 0.12852056],
       [ 0.05954058],
       [ 0.09295276],
       [ 0.01535029],
       [-0.00512814],
       [ 0.0703187 ],
       [-0.00405033],
       [-0.00081689],
       [-0.04392938],
       [ 0.02073935],
       [ 0.06061839],
       [-0.0105172 ],
       [-0.03315126],
       [-0.06548562],
       [ 0.0433734 ],
       [-0.06225218],
       [ 0.06385183],
       [ 0.03043966],
       [ 0.07247433],
       [-0.0191397 ],
       [-0.06656343],
       [-0.06009656],
       [ 0.06924089],
       [ 0.05954058],
       [-0.02668438],
       [-0.02021751],
       [-0.046085  ],
       [ 0.07139652],
       [-0.07949718],
       [ 0.00996123],
       [-0.03854032],
       [ 0.01966154],
       [ 0.02720622],
       [-0.00836158],
       [-0.01590626],
       [ 0.00457217],
       [-0.04285156],
       [ 0.00564998],
       [-0.03530688],
       [ 0.02397278],
       [-0.01806189],
       [ 0.04229559],
       [-0.0547075 ],
       [-0.00297252],
       [-0.06656343],
       [-0.01267283],
       [-0.04177375],
       [-0.03099563],
       [-0.00512814],
       [-0.05901875],
       [ 0.0250506 ],
       [-0.046085  ],
       [ 0.00349435],
       [ 0.05415152],
       [-0.04500719],
       [-0.05794093],
       [-0.05578531],
       [ 0.00133873],
       [ 0.03043966],
       [ 0.00672779],
       [ 0.04660684],
       [ 0.02612841],
       [ 0.04552903],
       [ 0.04013997],
       [-0.01806189],
       [ 0.01427248],
       [ 0.03690653],
       [ 0.00349435],
       [-0.07087468],
       [-0.03315126],
       [ 0.09403057],
       [ 0.03582872],
       [ 0.03151747],
       [-0.06548562],
       [-0.04177375],
       [-0.03961813],
       [-0.03854032],
       [-0.02560657],
       [-0.02345095],
       [-0.06656343],
       [ 0.03259528],
       [-0.046085  ],
       [-0.02991782],
       [-0.01267283],
       [-0.01590626],
       [ 0.07139652],
       [-0.03099563],
       [ 0.00026092],
       [ 0.03690653],
       [ 0.03906215],
       [-0.01482845],
       [ 0.00672779],
       [-0.06871905],
       [-0.00943939],
       [ 0.01966154],
       [ 0.07462995],
       [-0.00836158],
       [-0.02345095],
       [-0.046085  ],
       [ 0.05415152],
       [-0.03530688],
       [-0.03207344],
       [-0.0816528 ],
       [ 0.04768465],
       [ 0.06061839],
       [ 0.05630715],
       [ 0.09834182],
       [ 0.05954058],
       [ 0.03367309],
       [ 0.05630715],
       [-0.06548562],
       [ 0.16085492],
       [-0.05578531],
       [-0.02452876],
       [-0.03638469],
       [-0.00836158],
       [-0.04177375],
       [ 0.12744274],
       [-0.07734155],
       [ 0.02828403],
       [-0.02560657],
       [-0.06225218],
       [-0.00081689],
       [ 0.08864151],
       [-0.03207344],
       [ 0.03043966],
       [ 0.00888341],
       [ 0.00672779],
       [-0.02021751],
       [-0.02452876],
       [-0.01159501],
       [ 0.02612841],
       [-0.05901875],
       [-0.03638469],
       [-0.02452876],
       [ 0.01858372],
       [-0.0902753 ],
       [-0.00512814],
       [-0.05255187],
       [-0.02237314],
       [-0.02021751],
       [-0.0547075 ],
       [-0.00620595],
       [-0.01698407],
       [ 0.05522933],
       [ 0.07678558],
       [ 0.01858372],
       [-0.02237314],
       [ 0.09295276],
       [-0.03099563],
       [ 0.03906215],
       [-0.06117437],
       [-0.00836158],
       [-0.0374625 ],
       [-0.01375064],
       [ 0.07355214],
       [-0.02452876],
       [ 0.03367309],
       [ 0.0347509 ],
       [-0.03854032],
       [-0.03961813],
       [-0.00189471],
       [-0.03099563],
       [-0.046085  ],
       [ 0.00133873],
       [ 0.06492964],
       [ 0.04013997],
       [-0.02345095],
       [ 0.05307371],
       [ 0.04013997],
       [-0.02021751],
       [ 0.01427248],
       [-0.03422907],
       [ 0.00672779],
       [ 0.00457217],
       [ 0.03043966],
       [ 0.0519959 ],
       [ 0.06169621],
       [-0.00728377],
       [ 0.00564998],
       [ 0.05415152],
       [-0.00836158],
       [ 0.114509  ],
       [ 0.06708527],
       [-0.05578531],
       [ 0.03043966],
       [-0.02560657],
       [ 0.10480869],
       [-0.00620595],
       [-0.04716281],
       [-0.04824063],
       [ 0.08540807],
       [-0.01267283],
       [-0.03315126],
       [-0.00728377],
       [-0.01375064],
       [ 0.05954058],
       [ 0.02181716],
       [ 0.01858372],
       [-0.01159501],
       [-0.00297252],
       [ 0.01750591],
       [-0.02991782],
       [-0.02021751],
       [-0.05794093],
       [ 0.06061839],
       [-0.04069594],
       [-0.07195249],
       [-0.05578531],
       [ 0.04552903],
       [-0.00943939],
       [-0.03315126],
       [ 0.04984027],
       [-0.08488624],
       [ 0.00564998],
       [ 0.02073935],
       [-0.00728377],
       [ 0.10480869],
       [-0.02452876],
       [-0.00620595],
       [-0.03854032],
       [ 0.13714305],
       [ 0.17055523],
       [ 0.00241654],
       [ 0.03798434],
       [-0.05794093],
       [-0.00943939],
       [-0.02345095],
       [-0.0105172 ],
       [-0.03422907],
       [-0.00297252],
       [ 0.06816308],
       [ 0.00996123],
       [ 0.00241654],
       [-0.03854032],
       [ 0.02612841],
       [-0.08919748],
       [ 0.06061839],
       [-0.02884001],
       [-0.02991782],
       [-0.0191397 ],
       [-0.04069594],
       [ 0.01535029],
       [-0.02452876],
       [ 0.00133873],
       [ 0.06924089],
       [-0.06979687],
       [-0.02991782],
       [-0.046085  ],
       [ 0.01858372],
       [ 0.00133873],
       [-0.03099563],
       [-0.00405033],
       [ 0.01535029],
       [ 0.02289497],
       [ 0.04552903],
       [-0.04500719],
       [-0.03315126],
       [ 0.097264  ],
       [ 0.05415152],
       [ 0.12313149],
       [-0.08057499],
       [ 0.09295276],
       [-0.05039625],
       [-0.01159501],
       [-0.0277622 ],
       [ 0.05846277],
       [ 0.08540807],
       [-0.00081689],
       [ 0.00672779],
       [ 0.00888341],
       [ 0.08001901],
       [ 0.07139652],
       [-0.02452876],
       [-0.0547075 ],
       [-0.03638469],
       [ 0.0164281 ],
       [ 0.07786339],
       [-0.03961813],
       [ 0.01103904],
       [-0.04069594],
       [-0.03422907],
       [ 0.00564998],
       [ 0.08864151],
       [-0.03315126],
       [-0.05686312],
       [-0.03099563],
       [ 0.05522933],
       [-0.06009656],
       [ 0.00133873],
       [-0.02345095],
       [-0.07410811],
       [ 0.01966154],
       [-0.01590626],
       [-0.01590626],
       [ 0.03906215],
       [-0.0730303 ]])

In [10]:

    

diabetes_X.shape


    

    
Out[10]:





(442, 1)

In [11]:

    

diabetes.target.shape


    

    
Out[11]:





(442,)

In [12]:

    

regressor = linear_model.LinearRegression()
regressor.fit(diabetes_X, diabetes.target)


    

    
Out[12]:





LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)

In [13]:

    

import matplotlib.pyplot as plt
import seaborn


    

In [17]:

    

plt.scatter(diabetes_X, diabetes.target)
plt.plot(diabetes_X, regressor.predict(diabetes_X), color='red')
plt.show()


    

In [20]:

    

regressor.score(diabetes_X,diabetes.target)


    

    
Out[20]:





0.34392376022538029

In [21]:

    

np.mean((regressor.predict(diabetes_X) - diabetes.target)**2)


    

    
Out[21]:





3890.4565854612724

In [22]:

    

from sklearn.cross_validation import cross_val_score


    

In [23]:

    

scores = cross_val_score(regressor, diabetes_X, diabetes.target, cv=5)


    

In [24]:

    

scores


    

    
Out[24]:





array([ 0.20669771,  0.37001525,  0.38937267,  0.29361018,  0.36254055])

In [25]:

    

from sklearn import svm


    

In [61]:

    

svr_est = svm.SVR(C=1e3)


    

In [62]:

    

svr_est.fit(diabetes_X, diabetes.target)


    

    
Out[62]:





SVR(C=10000000000.0, cache_size=200, coef0=0.0, degree=3, epsilon=0.1,
  gamma='auto', kernel='rbf', max_iter=-1, shrinking=True, tol=0.001,
  verbose=False)

In [63]:

    

plt.scatter(diabetes_X,diabetes.target)
plt.plot(diabetes_X, svr_est.predict(diabetes_X), color='orange')
plt.show()


    

In [29]:

    

svr_est.score(diabetes_X, diabetes.target
            )


    

    
Out[29]:





-0.011761447401173353

In [30]:

    

cross_val_score(svr_est, diabetes_X, diabetes.target, cv=5)


    

    
Out[30]:





array([-0.00817555, -0.07854939, -0.00658466, -0.05683521, -0.02648136])

In [31]:

    

svr_lin_est = svm.SVR(kernel='linear')
svr_lin_est.fit(diabetes_X, diabetes.target)


    

    
Out[31]:





SVR(C=1.0, cache_size=200, coef0=0.0, degree=3, epsilon=0.1, gamma='auto',
  kernel='linear', max_iter=-1, shrinking=True, tol=0.001, verbose=False)

In [33]:

    

plt.scatter(diabetes_X,diabetes.target)
plt.plot(diabetes_X,svr_lin_est.predict(diabetes_X),color='orange')
plt.show()


    

In [34]:

    

np.sin(np.pi)


    

    
Out[34]:





1.2246467991473532e-16

In [38]:

    

xaxis = np.linspace(-np.pi, np.pi,20)
plt.scatter(xaxis, np.sin(xaxis))
plt.show()


    

In [39]:

    

np.sin(xaxis)


    

    
Out[39]:





array([ -1.22464680e-16,  -3.24699469e-01,  -6.14212713e-01,
        -8.37166478e-01,  -9.69400266e-01,  -9.96584493e-01,
        -9.15773327e-01,  -7.35723911e-01,  -4.75947393e-01,
        -1.64594590e-01,   1.64594590e-01,   4.75947393e-01,
         7.35723911e-01,   9.15773327e-01,   9.96584493e-01,
         9.69400266e-01,   8.37166478e-01,   6.14212713e-01,
         3.24699469e-01,   1.22464680e-16])

In [48]:

    

svr_est_sin = svm.SVR()
svr_est_sin.fit(xaxis.reshape(-1,1), np.sin(xaxis))


    

    
Out[48]:





SVR(C=1.0, cache_size=200, coef0=0.0, degree=3, epsilon=0.1, gamma='auto',
  kernel='rbf', max_iter=-1, shrinking=True, tol=0.001, verbose=False)

In [55]:

    

plt.scatter(xaxis, svr_est_sin.predict(xaxis.reshape(-1,1)))
plt.plot(np.linspace(-np.pi,np.pi,50), np.sin(np.linspace(-np.pi,np.pi,50)))
plt.show()


    

In [57]:

    

plt.scatter(np.linspace(-2*np.pi,2*np.pi,50), svr_est_sin.predict(np.linspace(-2*np.pi,2*np.pi,50).reshape(-1,1)))
plt.plot(np.linspace(-2*np.pi,2*np.pi,50), np.sin(np.linspace(-2*np.pi,2*np.pi,50)))
plt.show()


    

In [ ]: