吴恩达机器学习课程笔记+代码实现(18)Python实现正则化的线性回归和偏差/方差(Programming Exercise 5)

Programming Exercise 5:Regularized Linear Regression and Bias v.s.Variance

Python版本3.6
编译环境：anaconda Jupyter Notebook
链接：实验数据和实验指导书
提取码：i7co
本章课程笔记部分见：应用机器学习的建议(Advice for Applying Machine Learning)
在本练习中，我们要实现正则化的线性回归，并且使用它来学习在不同的方差和偏差性能的模型

%matplotlib inline
#IPython的内置magic函数，可以省掉plt.show()，在其他IDE中是不会支持的
import numpy as np
import pandas as pd
import matplotlib as mpl
import matplotlib.pyplot as plt
import seaborn as sns
sns.set(style="whitegrid",color_codes=True)
import scipy.io as sio
import scipy.optimize as opt
from sklearn.metrics import classification_report#这个包是评价报告

加载数据集和可视化

data = sio.loadmat('ex5data1.mat')
data

{'__header__': b'MATLAB 5.0 MAT-file, Platform: GLNXA64, Created on: Fri Nov  4 22:27:26 2011',
 '__version__': '1.0',
 '__globals__': [],
 'X': array([[-15.93675813],
        [-29.15297922],
        [ 36.18954863],
        [ 37.49218733],
        [-48.05882945],
        [ -8.94145794],
        [ 15.30779289],
        [-34.70626581],
        [  1.38915437],
        [-44.38375985],
        [  7.01350208],
        [ 22.76274892]]),
 'y': array([[ 2.13431051],
        [ 1.17325668],
        [34.35910918],
        [36.83795516],
        [ 2.80896507],
        [ 2.12107248],
        [14.71026831],
        [ 2.61418439],
        [ 3.74017167],
        [ 3.73169131],
        [ 7.62765885],
        [22.7524283 ]]),
 'Xtest': array([[-33.31800399],
        [-37.91216403],
        [-51.20693795],
        [ -6.13259585],
        [ 21.26118327],
        [-40.31952949],
        [-14.54153167],
        [ 32.55976024],
        [ 13.39343255],
        [ 44.20988595],
        [ -1.14267768],
        [-12.76686065],
        [ 34.05450539],
        [ 39.22350028],
        [  1.97449674],
        [ 29.6217551 ],
        [-23.66962971],
        [ -9.01180139],
        [-55.94057091],
        [-35.70859752],
        [  9.51020533]]),
 'ytest': array([[ 3.31688953],
        [ 5.39768952],
        [ 0.13042984],
        [ 6.1925982 ],
        [17.08848712],
        [ 0.79950805],
        [ 2.82479183],
        [28.62123334],
        [17.04639081],
        [55.38437334],
        [ 4.07936733],
        [ 8.27039793],
        [31.32355102],
        [39.15906103],
        [ 8.08727989],
        [24.11134389],
        [ 2.4773548 ],
        [ 6.56606472],
        [ 6.0380888 ],
        [ 4.69273956],
        [10.83004606]]),
 'Xval': array([[-16.74653578],
        [-14.57747075],
        [ 34.51575866],
        [-47.01007574],
        [ 36.97511905],
        [-40.68611002],
        [ -4.47201098],
        [ 26.53363489],
        [-42.7976831 ],
        [ 25.37409938],
        [-31.10955398],
        [ 27.31176864],
        [ -3.26386201],
        [ -1.81827649],
        [-40.7196624 ],
        [-50.01324365],
        [-17.41177155],
        [  3.5881937 ],
        [  7.08548026],
        [ 46.28236902],
        [ 14.61228909]]),
 'yval': array([[ 4.17020201e+00],
        [ 4.06726280e+00],
        [ 3.18730676e+01],
        [ 1.06236562e+01],
        [ 3.18360213e+01],
        [ 4.95936972e+00],
        [ 4.45159880e+00],
        [ 2.22763185e+01],
        [-4.38738274e-05],
        [ 2.05038016e+01],
        [ 3.85834476e+00],
        [ 1.93650529e+01],
        [ 4.88376281e+00],
        [ 1.10971588e+01],
        [ 7.46170827e+00],
        [ 1.47693464e+00],
        [ 2.71916388e+00],
        [ 1.09269007e+01],
        [ 8.34871235e+00],
        [ 5.27819280e+01],
        [ 1.33573396e+01]])}

def load_data():
    """for ex5
    d['X'] shape = (12, 1)
    pandas has trouble taking this 2d ndarray to construct a dataframe, so I ravel
    the results
    """
    d = sio.loadmat('ex5data1.mat')
    return map(np.ravel, [d['X'], d['y'], d['Xval'], d['yval'], d['Xtest'], d['ytest']])

X, y, Xval, yval, Xtest, ytest = load_data()

df = pd.DataFrame({
   'water_level':X, 'flow':y})

sns.lmplot('water_level', 'flow', data=df, fit_reg    =False, size=5)

<seaborn.axisgrid.FacetGrid at 0xd59b2e9400>

X, Xval, Xtest = [np.insert(x.reshape(x.shape[0], 1), 0, np.ones(x.shape[0]), axis=1) for x in (X, Xval, Xtest)]

代价函数

def cost(theta, X