第9章 逻辑回归

 

9-1 什么是逻辑回归

 

 

 

 

 

 

 

 

 

 

 

 

Notbook 示例

 

 

 Notbook 源码

复制代码
 1 Sigmoid 函数
 2 [1]
 3 import numpy as np
 4 import matplotlib.pyplot as plt
 5 [2]
 6 def sigmoid(t):
 7     return 1 / (1 + np.exp(-t))
 8 [3]
 9 X = np.linspace(-10,10,500)
10 y = sigmoid(X)
11 
12 plt.plot(X,y)
13 [<matplotlib.lines.Line2D at 0x18b21371c40>]
复制代码

 

9-2 逻辑回归的损失函数

 

 

 

 

 

 

 

 

 

 

9-3 逻辑回归损失函数的梯度

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 9-4 and 9-5 实现逻辑回归算法与决策边界

 

 

 

 

 

 

 

 

Notbook 示例

 

 

Notbook 源码

复制代码
  1 实现逻辑回归
  2 [1]
  3 import numpy as np 
  4 import matplotlib.pyplot as plt
  5 from sklearn import datasets
  6 
  7 iris = datasets.load_iris()
  8 [2]
  9 X = iris.data
 10 y = iris.target
 11 [3]
 12 X = X[y<2, :2]
 13 y = y[y<2]
 14 [4]
 15 X.shape
 16 (100, 2)
 17 [5]
 18 y.shape
 19 (100,)
 20 [6]
 21 plt.scatter(X[y==0,0], X[y==0,1],color="red")
 22 plt.scatter(X[y==1,0], X[y==1,1],color="blue")
 23 <matplotlib.collections.PathCollection at 0x18d15300d00>
 24 
 25 使用逻辑回归
 26 [7]
 27 from playML.model_selection import train_test_split
 28 
 29 X_train, X_test, y_train, y_test = train_test_split(X, y, seed=666)
 30 [8]
 31 from playML.LogistcRegression import LogisticRegression
 32 
 33 log_reg = LogisticRegression()
 34 log_reg.fit(X_train, y_train)
 35 LogisticRegression()
 36 [9]
 37 log_reg.score(X_test, y_test)
 38 1.0
 39 [10]
 40 log_reg.predict_proba(X_test)
 41 array([9.62648257e-01, 9.95348774e-01, 1.00823761e-01, 6.17589825e-03,
 42        1.67472308e-02, 6.93067583e-03, 2.43033667e-02, 9.99216051e-01,
 43        9.92621004e-01, 7.93937643e-01, 2.16979735e-02, 8.39808132e-04,
 44        2.12177556e-01, 1.67472308e-02, 8.93306850e-01, 8.47220912e-01,
 45        8.78361232e-01, 2.82569244e-01, 3.56919294e-02, 1.55520124e-01])
 46 [11]
 47 y_test
 48 array([1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0])
 49 [12]
 50 log_reg.predict(X_test)
 51 array([1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0])
 52 决策边界
 53 [13]
 54 log_reg.coef_
 55 array([ 3.80096484, -6.28176101])
 56 [14]
 57 log_reg.interception_
 58 -1.0912939737572598
 59 [15]
 60 def X2(X1):
 61     return (-log_reg.coef_[0] * X1 - log_reg.interception_) / log_reg.coef_[1]
 62 [16]
 63 X1_plot = np.linspace(4, 8,1000)
 64 X2_plot = X2(X1_plot)
 65 [17]
 66 plt.scatter(X[y==0,0], X[y==0,1],color="red")
 67 plt.scatter(X[y==1,0], X[y==1,1],color="blue")
 68 plt.plot(X1_plot,X2_plot)
 69 [<matplotlib.lines.Line2D at 0x18d15ac2fd0>]
 70 
 71 [18]
 72 plt.scatter(X_test[y_test==0,0], X_test[y_test==0,1],color="red")
 73 plt.scatter(X_test[y_test==1,0], X_test[y_test==1,1],color="blue")
 74 plt.plot(X1_plot,X2_plot)
 75 [<matplotlib.lines.Line2D at 0x18d15b4d2e0>]
 76 
 77 [19]
 78 # plot_decision_boundary()函数：绘制模型在二维特征空间的决策边界；
 79 def plot_decision_boundary(model, axis):
 80     # model：算法模型；
 81     # axis：区域坐标轴的范围，其中 0,1,2,3 分别对应 x 轴和 y 轴的范围；
 82     
 83     # 1）将坐标轴等分为无数的小点，将 x、y 轴分别等分 （坐标轴范围最大值 - 坐标轴范围最小值）*100 份，
 84     # np.meshgrid()：
 85     x0, x1 = np.meshgrid(
 86         np.linspace(axis[0], axis[1], int((axis[1]-axis[0])*100)).reshape(-1,1),
 87         np.linspace(axis[2], axis[3], int((axis[3]-axis[2])*100)).reshape(-1,1)
 88     )
 89     # np.c_()：
 90     X_new = np.c_[x0.ravel(), x1.ravel()]
 91     
 92     # 2）model.predict(X_new)：将分割出的所有的点，都使用模型预测
 93     y_predict = model.predict(X_new)
 94     zz = y_predict.reshape(x0.shape)
 95     
 96     # 3）绘制预测结果
 97     from matplotlib.colors import ListedColormap
 98     custom_cmap = ListedColormap(['#EF9A9A','#FFF59D','#90CAF9'])
 99     
100     plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
101 [20]
102 plot_decision_boundary(log_reg, axis=[4, 7.5, 1.5, 4.5])
103 plt.scatter(X[y==0, 0], X[y==0, 1] )
104 plt.scatter(X[y==1, 0], X[y==1, 1])
105 C:\Users\Administrator\AppData\Local\Temp\ipykernel_12716\2692106441.py:23: UserWarning: The following kwargs were not used by contour: 'linewidth'
106   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
107 
108 <matplotlib.collections.PathCollection at 0x18d15d601c0>
109 
110 kNN的决策边界
111 [21]
112 from sklearn.neighbors import KNeighborsClassifier
113 
114 knn_clf = KNeighborsClassifier()
115 knn_clf.fit(X_train,y_train)
116 KNeighborsClassifier()
117 [22]
118 knn_clf.score(X_test,y_test)
119 1.0
120 [23]
121 plot_decision_boundary(knn_clf, axis=[4, 7.5, 1.5, 4.5])
122 plt.scatter(X[y==0, 0], X[y==0, 1] )
123 plt.scatter(X[y==1, 0], X[y==1, 1])
124 C:\Users\Administrator\AppData\Local\Temp\ipykernel_12716\2692106441.py:23: UserWarning: The following kwargs were not used by contour: 'linewidth'
125   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
126 
127 <matplotlib.collections.PathCollection at 0x18d15f68d90>
128 
129 [24]
130 knn_clf_all = KNeighborsClassifier()
131 knn_clf_all.fit(iris.data[:,:2],iris.target)
132 KNeighborsClassifier()
133 [25]
134 plot_decision_boundary(knn_clf_all, axis=[4, 8, 1.5, 4.5])
135 plt.scatter(iris.data[iris.target==0,0],iris.data[iris.target==0,1])
136 plt.scatter(iris.data[iris.target==1,0],iris.data[iris.target==1,1])
137 plt.scatter(iris.data[iris.target==2,0],iris.data[iris.target==2,1])
138 C:\Users\Administrator\AppData\Local\Temp\ipykernel_12716\2692106441.py:23: UserWarning: The following kwargs were not used by contour: 'linewidth'
139   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
140 
141 <matplotlib.collections.PathCollection at 0x18d15fd9a30>
142 
143 [26]
144 knn_clf_all = KNeighborsClassifier(n_neighbors=50)
145 knn_clf_all.fit(iris.data[:,:2],iris.target)
146 KNeighborsClassifier(n_neighbors=50)
147 [27]
148 plot_decision_boundary(knn_clf_all, axis=[4, 8, 1.5, 4.5])
149 plt.scatter(iris.data[iris.target==0,0],iris.data[iris.target==0,1])
150 plt.scatter(iris.data[iris.target==1,0],iris.data[iris.target==1,1])
151 plt.scatter(iris.data[iris.target==2,0],iris.data[iris.target==2,1])
152 C:\Users\Administrator\AppData\Local\Temp\ipykernel_12716\2692106441.py:23: UserWarning: The following kwargs were not used by contour: 'linewidth'
153   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
154 
155 <matplotlib.collections.PathCollection at 0x18d1606c8b0>
复制代码

 

9-6 在逻辑回归中使用多项式特征

 

 

 

 Notbook 示例

 

 

 

 

 

 

Notbook 源码

复制代码
 1 逻辑回归中添加多项式特征
 2 [1]
 3 import numpy as np
 4 import matplotlib.pyplot as plt
 5 [2]
 6 np.random.seed(666)
 7 X = np.random.normal(0,1,size=(200, 2))
 8 y = np.array(X[:,0]**2 + X[:,1]**2 < 1.5 ,dtype='int')
 9 [3]
10 plt.scatter(X[y==0,0], X[y==0,1])
11 plt.scatter(X[y==1,0], X[y==1,1])
12 <matplotlib.collections.PathCollection at 0x22a90fd94f0>
13 
14 使用逻辑回归
15 [4]
16 from playML.LogistcRegression import LogisticRegression
17 [5]
18 log_reg = LogisticRegression()
19 log_reg.fit(X,y)
20 LogisticRegression()
21 [6]
22 log_reg.score(X,y)
23 0.61
24 [7]
25 def plot_decision_boundary(model, axis):
26 
27     x0, x1 = np.meshgrid(
28         np.linspace(axis[0], axis[1], int((axis[1]-axis[0])*100)).reshape(-1,1),
29         np.linspace(axis[2], axis[3], int((axis[3]-axis[2])*100)).reshape(-1,1)
30     )
31     X_new = np.c_[x0.ravel(), x1.ravel()]
32     
33     y_predict = model.predict(X_new)
34     zz = y_predict.reshape(x0.shape)
35 
36     from matplotlib.colors import ListedColormap
37     custom_cmap = ListedColormap(['#EF9A9A','#FFF59D','#90CAF9'])
38     
39     plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
40 [8]
41 plot_decision_boundary(log_reg, axis=[-4,4,-4,4])
42 plt.scatter(X[y==0,0], X[y==0,1])
43 plt.scatter(X[y==1,0], X[y==1,1])
44 C:\Users\Administrator\AppData\Local\Temp\ipykernel_9172\3130018029.py:15: UserWarning: The following kwargs were not used by contour: 'linewidth'
45   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
46 
47 <matplotlib.collections.PathCollection at 0x22a92dbf580>
48 
49 [9]
50 from sklearn.pipeline import Pipeline
51 from sklearn.preprocessing import PolynomialFeatures
52 from sklearn.preprocessing import StandardScaler
53 
54 def PolynomialLogisticRegression(degree):
55     return Pipeline([
56     ("poly",PolynomialFeatures(degree=degree)),
57     ("std_scaler",StandardScaler()),
58     ("log_reg",LogisticRegression())
59 ])
60 [10]
61 poly_log_reg = PolynomialLogisticRegression(degree=2)
62 poly_log_reg.fit(X,y)
63 Pipeline(steps=[('poly', PolynomialFeatures()),
64                 ('std_scaler', StandardScaler()),
65                 ('log_reg', LogisticRegression())])
66 [11]
67 poly_log_reg.score(X,y)
68 0.97
69 [12]
70 plot_decision_boundary(poly_log_reg, axis=[-4,4,-4,4])
71 plt.scatter(X[y==0,0], X[y==0,1])
72 plt.scatter(X[y==1,0], X[y==1,1])
73 C:\Users\Administrator\AppData\Local\Temp\ipykernel_9172\3130018029.py:15: UserWarning: The following kwargs were not used by contour: 'linewidth'
74   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
75 
76 <matplotlib.collections.PathCollection at 0x22a92aca6d0>
77 
78 [13]
79 poly_log_reg2 = PolynomialLogisticRegression(degree=20)
80 poly_log_reg2.fit(X,y)
81 Pipeline(steps=[('poly', PolynomialFeatures(degree=20)),
82                 ('std_scaler', StandardScaler()),
83                 ('log_reg', LogisticRegression())])
84 [14]
85 plot_decision_boundary(poly_log_reg2, axis=[-4,4,-4,4])
86 plt.scatter(X[y==0,0], X[y==0,1])
87 plt.scatter(X[y==1,0], X[y==1,1])
88 C:\Users\Administrator\AppData\Local\Temp\ipykernel_9172\3130018029.py:15: UserWarning: The following kwargs were not used by contour: 'linewidth'
89   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
90 
91 <matplotlib.collections.PathCollection at 0x22a92ba6cd0>
复制代码

 

 9-7 scikit-learn中的逻辑回归

 

 

 

Notbook 示例

 

 Notbook 源码

复制代码
  1 [1]
  2 import numpy as np
  3 import matplotlib.pyplot as plt
  4 [2]
  5 np.random.seed(666)
  6 X = np.random.normal(0,1,size=(200, 2))
  7 y = np.array(X[:,0]**2 + X[:,1] < 1.5 ,dtype='int')
  8 for _ in range(20):
  9     y[np.random.randint(200)] = 1
 10 [3]
 11 plt.scatter(X[y==0,0], X[y==0,1])
 12 plt.scatter(X[y==1,0], X[y==1,1])
 13 <matplotlib.collections.PathCollection at 0x2397e35b220>
 14 
 15 [4]
 16 from sklearn.model_selection import train_test_split
 17 X_train, X_test, y_train, y_test = train_test_split(X, y,random_state=666)
 18 使用scikit-learn 中的逻辑回归
 19 [5]
 20 from sklearn.linear_model import LogisticRegression
 21 
 22 log_reg = LogisticRegression()
 23 log_reg.fit(X_train, y_train)
 24 LogisticRegression()
 25 [6]
 26 log_reg.score(X_train,y_train)
 27 0.7933333333333333
 28 [7]
 29 log_reg.score(X_test,y_test)
 30 0.86
 31 [8]
 32 def plot_decision_boundary(model, axis):
 33 
 34     x0, x1 = np.meshgrid(
 35         np.linspace(axis[0], axis[1], int((axis[1]-axis[0])*100)).reshape(-1,1),
 36         np.linspace(axis[2], axis[3], int((axis[3]-axis[2])*100)).reshape(-1,1)
 37     )
 38     X_new = np.c_[x0.ravel(), x1.ravel()]
 39     
 40     y_predict = model.predict(X_new)
 41     zz = y_predict.reshape(x0.shape)
 42 
 43     from matplotlib.colors import ListedColormap
 44     custom_cmap = ListedColormap(['#EF9A9A','#FFF59D','#90CAF9'])
 45     
 46     plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
 47 [9]
 48 plot_decision_boundary(log_reg, axis=[-4,4,-4,4])
 49 plt.scatter(X[y==0,0], X[y==0,1])
 50 plt.scatter(X[y==1,0], X[y==1,1])
 51 C:\Users\Administrator\AppData\Local\Temp\ipykernel_13668\3130018029.py:15: UserWarning: The following kwargs were not used by contour: 'linewidth'
 52   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
 53 
 54 <matplotlib.collections.PathCollection at 0x23903dc3460>
 55 
 56 [10]
 57 from sklearn.pipeline import Pipeline
 58 from sklearn.preprocessing import PolynomialFeatures
 59 from sklearn.preprocessing import StandardScaler
 60 
 61 def PolynomialLogisticRegression(degree):
 62     return Pipeline([
 63     ("poly",PolynomialFeatures(degree=degree)),
 64     ("std_scaler",StandardScaler()),
 65     ("log_reg",LogisticRegression())
 66 ])
 67 [11]
 68 poly_log_reg = PolynomialLogisticRegression(degree=2)
 69 poly_log_reg.fit(X_train,y_train)
 70 Pipeline(steps=[('poly', PolynomialFeatures()),
 71                 ('std_scaler', StandardScaler()),
 72                 ('log_reg', LogisticRegression())])
 73 [12]
 74 poly_log_reg.score(X_train,y_train)
 75 0.9066666666666666
 76 [13]
 77 poly_log_reg.score(X_test,y_test)
 78 0.94
 79 [14]
 80 plot_decision_boundary(poly_log_reg, axis=[-4,4,-4,4])
 81 plt.scatter(X[y==0,0], X[y==0,1])
 82 plt.scatter(X[y==1,0], X[y==1,1])
 83 C:\Users\Administrator\AppData\Local\Temp\ipykernel_13668\3130018029.py:15: UserWarning: The following kwargs were not used by contour: 'linewidth'
 84   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
 85 
 86 <matplotlib.collections.PathCollection at 0x23902a64790>
 87 
 88 [15]
 89 poly_log_reg2 = PolynomialLogisticRegression(degree=20)
 90 poly_log_reg2.fit(X_train,y_train)
 91 Pipeline(steps=[('poly', PolynomialFeatures(degree=20)),
 92                 ('std_scaler', StandardScaler()),
 93                 ('log_reg', LogisticRegression())])
 94 [16]
 95 poly_log_reg2.score(X_train,y_train)
 96 0.94
 97 [17]
 98 poly_log_reg2.score(X_test,y_test)
 99 0.92
100 [18]
101 plot_decision_boundary(poly_log_reg2, axis=[-4,4,-4,4])
102 plt.scatter(X[y==0,0], X[y==0,1])
103 plt.scatter(X[y==1,0], X[y==1,1])
104 C:\Users\Administrator\AppData\Local\Temp\ipykernel_13668\3130018029.py:15: UserWarning: The following kwargs were not used by contour: 'linewidth'
105   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
106 
107 <matplotlib.collections.PathCollection at 0x239034965b0>
108 
109 [19]
110 def PolynomialLogisticRegression(degree, C):
111     return Pipeline([
112     ("poly",PolynomialFeatures(degree=degree)),
113     ("std_scaler",StandardScaler()),
114     ("log_reg",LogisticRegression(C=C))
115 ])
116 [20]
117 poly_log_reg3 = PolynomialLogisticRegression(degree=20, C=0.1)
118 poly_log_reg3.fit(X_train,y_train)
119 Pipeline(steps=[('poly', PolynomialFeatures(degree=20)),
120                 ('std_scaler', StandardScaler()),
121                 ('log_reg', LogisticRegression(C=0.1))])
122 [21]
123 poly_log_reg3.score(X_train,y_train)
124 0.84
125 [22]
126 poly_log_reg3.score(X_test,y_test)
127 0.92
128 [23]
129 plot_decision_boundary(poly_log_reg3, axis=[-4,4,-4,4])
130 plt.scatter(X[y==0,0], X[y==0,1])
131 plt.scatter(X[y==1,0], X[y==1,1])
132 C:\Users\Administrator\AppData\Local\Temp\ipykernel_13668\3130018029.py:15: UserWarning: The following kwargs were not used by contour: 'linewidth'
133   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
134 
135 <matplotlib.collections.PathCollection at 0x23902feb970>
136 
137 [24]
138 def PolynomialLogisticRegression(degree, C, penalty='l2'):
139     return Pipeline([
140     ("poly",PolynomialFeatures(degree=degree)),
141     ("std_scaler",StandardScaler()),
142     ("log_reg",LogisticRegression(C=C))
143 ])
144 [25]
145 poly_log_reg4 = PolynomialLogisticRegression(degree=20,C=0.1,penalty='l1')
146 poly_log_reg4.fit(X_train,y_train)
147 Pipeline(steps=[('poly', PolynomialFeatures(degree=20)),
148                 ('std_scaler', StandardScaler()),
149                 ('log_reg', LogisticRegression(C=0.1))])
150 [26]
151 poly_log_reg4.score(X_train,y_train)
152 0.84
153 [27]
154 poly_log_reg4.score(X_test,y_test)
155 0.92
156 [28]
157 plot_decision_boundary(poly_log_reg4, axis=[-4,4,-4,4])
158 plt.scatter(X[y==0,0], X[y==0,1])
159 plt.scatter(X[y==1,0], X[y==1,1])
160 C:\Users\Administrator\AppData\Local\Temp\ipykernel_13668\3130018029.py:15: UserWarning: The following kwargs were not used by contour: 'linewidth'
161   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
162 
163 <matplotlib.collections.PathCollection at 0x23902b3c760>
复制代码

 

9-8 OvR与OvO

 

 

 

 

 

 Notbook 示例

 

 

Notbook 源码

复制代码
 1 OvR 和 OvO
 2 [1]
 3 import numpy as np
 4 import matplotlib.pyplot as plt
 5 from sklearn import datasets
 6 
 7 iris = datasets.load_iris()
 8 X = iris.data[:, :2]
 9 y = iris.target
10 [2]
11 from sklearn.model_selection import train_test_split
12 
13 X_train, X_test, y_train, y_test = train_test_split(X, y,random_state=7)
14 [3]
15 from sklearn.linear_model import LogisticRegression
16 
17 log_reg = LogisticRegression(multi_class="ovr")
18 log_reg.fit(X_train,y_train)
19 LogisticRegression(multi_class='ovr')
20 [4]
21 log_reg.score(X_test,y_test)
22 0.6052631578947368
23 [5]
24 def plot_decision_boundary(model, axis):
25 
26     x0, x1 = np.meshgrid(
27         np.linspace(axis[0], axis[1], int((axis[1]-axis[0])*100)).reshape(-1,1),
28         np.linspace(axis[2], axis[3], int((axis[3]-axis[2])*100)).reshape(-1,1)
29     )
30     X_new = np.c_[x0.ravel(), x1.ravel()]
31     
32     y_predict = model.predict(X_new)
33     zz = y_predict.reshape(x0.shape)
34 
35     from matplotlib.colors import ListedColormap
36     custom_cmap = ListedColormap(['#EF9A9A','#FFF59D','#90CAF9'])
37     
38     plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
39 [6]
40 plot_decision_boundary(log_reg, axis=[4,8.5,1.5,4.5])
41 plt.scatter(X[y==0,0], X[y==0,1])
42 plt.scatter(X[y==1,0], X[y==1,1])
43 plt.scatter(X[y==2,0], X[y==2,1])
44 C:\Users\Administrator\AppData\Local\Temp\ipykernel_8836\3130018029.py:15: UserWarning: The following kwargs were not used by contour: 'linewidth'
45   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
46 
47 <matplotlib.collections.PathCollection at 0x277418551f0>
48 
49 [7]
50 log_reg2 = LogisticRegression(multi_class="multinomial", solver='newton-cg')
51 [8]
52 log_reg2.fit(X_train, y_train)
53 log_reg2.score(X_test,y_test)
54 0.6842105263157895
55 [9]
56 plot_decision_boundary(log_reg2, axis=[4,8.5,1.5,4.5])
57 plt.scatter(X[y==0,0], X[y==0,1])
58 plt.scatter(X[y==1,0], X[y==1,1])
59 plt.scatter(X[y==2,0], X[y==2,1])
60 C:\Users\Administrator\AppData\Local\Temp\ipykernel_8836\3130018029.py:15: UserWarning: The following kwargs were not used by contour: 'linewidth'
61   plt.contourf(x0, x1, zz, linewidth=5, cmap=custom_cmap)
62 
63 <matplotlib.collections.PathCollection at 0x27741808880>
64 
65 使用所有数据
66 [10]
67 X = iris.data
68 y = iris.target
69 
70 from sklearn.model_selection import train_test_split
71 
72 X_train, X_test, y_train, y_test = train_test_split(X, y,random_state=654)
73 [11]
74 log_reg = LogisticRegression(multi_class="ovr")
75 log_reg.fit(X_train, y_train)
76 log_reg.score(X_test,y_test)
77 0.9473684210526315
78 [12]
79 log_reg2 = LogisticRegression()
80 log_reg2.fit(X_train, y_train)
81 log_reg2.score(X_test,y_test)
82 0.9736842105263158
83 OvO and OvR
84 [13]
85 from sklearn.multiclass import OneVsRestClassifier
86 
87 ovr =OneVsRestClassifier(log_reg)
88 ovr.fit(X_train,y_train)
89 ovr.score(X_test,y_test)
90 0.9473684210526315
91 [14]
92 from sklearn.multiclass import OneVsOneClassifier
93 
94 ovo = OneVsOneClassifier(log_reg)
95 ovo.fit(X_train,y_train)
96 ovo.score(X_test,y_test)
97 0.9736842105263158
复制代码

 

posted @   Cai-Gbro  阅读(73)  评论(0)  编辑  收藏  举报
相关博文：
·  第8章上 多项式回归与模型泛化
·  第8章下 多项式回归与模型泛化
·  chapter3——逻辑回归手动+sklean版本
·  逻辑回归模型
·  逻辑回归练习
阅读排行：
· TypeScript + Deepseek 打造卜卦网站：技术与玄学的结合
· Manus的开源复刻OpenManus初探
· AI 智能体引爆开源社区「GitHub 热点速览」
· 三行代码完成国际化适配，妙~啊~
· .NET Core 中如何实现缓存的预热？
点击右上角即可分享
微信分享提示
SQL AI 助手