利用Kneighbors构建乳腺癌数据分类器

Kneighbors分类算法是一种十分简单的分类算法，他通过距离目标点最近的n个点的类别通过“投票”的方式来决定目标点的类别。

导入机器学习可能用到的包

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import sklearn
%matplotlib inline

导入分类器、训练集与测试集拆分函数以及乳腺癌数据

from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.datasets import load_breast_cancer

拆分数据为训练集与测试集，其中stratify参数为拆分时按类别比例（y）拆分

cancer = load_breast_cancer()
X_train,X_test,y_train,y_test = train_test_split(cancer.data,cancer.target,stratify = cancer.target)

构建分类器并评分、绘图，其中score所得为可决系数，即回归平方和与总偏差平方和之比。

Test_score = []
Train_score = []
for i in range(1,10):
    knn = KNeighborsClassifier(n_neighbors=i).fit(X_train,y_train)
    Test_score.append(knn.score(X_test,y_test))
    Train_score.append(knn.score(X_train,y_train))
plt.plot(range(1,10),Test_score,label = 'Test_score')
plt.plot(range(1,10),Train_score,label = 'Train_score')
plt.xlabel('Number of neighbors')
plt.ylabel('Accuracy')
plt.legend()

<matplotlib.legend.Legend at 0x1e34ae12160>