逻辑回归初学2

逻辑回归初学

老规矩，导入库

from sklearn.linear_model import LogisticRegression as LR
import numpy as np
from matplotlib import pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
from sklearn.datasets import load_breast_cancer

max_iter学习曲线

l2 = []
l2test = []
cancer = load_breast_cancer()
x = cancer.data
y = cancer.target

x_train,x_test,y_train,y_test = train_test_split(x,y,test_size=0.3)

for i in np.arange(1,201,10):
    L2 = LR(penalty='l2',solver = 'liblinear', C=0.8,max_iter=i)
    L2 = L2.fit(x_train,y_train)
    l2.append(accuracy_score(L2.predict(x_train),y_train))
    l2test.append(accuracy_score(L2.predict(x_test),y_test))

graph = [l2,l2test]
color = ['black','gray']
label = ['l2','l2test']

plt.figure(figsize=(8,6))
for i in range(len(graph)):
    plt.plot(np.arange(1,201,10),graph[i],color[i],label=label[i])
plt.legend()
plt.xticks(np.arange(1,201,10))
plt.show()

使用n_iter_来看本次求解中真正实现的迭代次数

L2 = LR(penalty='l2',solver = 'liblinear', C=0.8,max_iter=300).fit(x_train,y_train)
L2.n_iter_

array([21], dtype=int32)

降维

from sklearn.model_selection import cross_val_score
from sklearn.feature_selection import SelectFromModel

cancer.data.shape

(569, 30)

Cancer_LR = LR(solver = 'liblinear', C=0.8)

交叉验证

cross_val_score(Cancer_LR,cancer.data,cancer.target,cv=10).mean()

0.9508998790078644

x_embedded = SelectFromModel(Cancer_LR).fit_transform(cancer.data,cancer.target)
x_embedded.shape

(569, 9)

cross_val_score(Cancer_LR,x_embedded,cancer.target,cv=10).mean()

0.9368323826808401

调整C的学习曲线来降维

fullx = []
delx = []

c = np.arange(0.01,10.01,0.5)

for i in c:
    Cancer_LR = LR(solver = 'liblinear', C=i)
    fullx.append(cross_val_score(Cancer_LR,cancer.data,cancer.target,cv=10).mean())
    x_embedded = SelectFromModel(Cancer_LR).fit_transform(cancer.data,cancer.target)
    delx.append(cross_val_score(Cancer_LR,x_embedded,cancer.target,cv=10).mean())

print(max(delx),c[delx.index(max(delx))])
plt.figure(figsize = (20,8))
plt.plot(c,fullx,label="full")
plt.plot(c,delx,label="feature selection")
plt.xticks(c)
plt.legend()
plt.show()

fullx = []
delx = []
c = np.arange(6.05,7.05,0.005)

for i in c:
    Cancer_LR = LR(solver = 'liblinear', C=i)
    fullx.append(cross_val_score(Cancer_LR,cancer.data,cancer.target,cv=10).mean())
    x_embedded = SelectFromModel(Cancer_LR).fit_transform(cancer.data,cancer.target)
    delx.append(cross_val_score(Cancer_LR,x_embedded,cancer.target,cv=10).mean())

print(max(delx),c[delx.index(max(delx))])
plt.figure(figsize = (20,8))
plt.plot(c,fullx,label="full")
plt.plot(c,delx,label="feature selection")
plt.xticks(c)
plt.legend()
plt.show()