K邻近分类算法

# -*- coding: utf-8 -*-
"""
Created on Thu Jun 28 17:16:19 2018

@author: zhen
"""
from sklearn.model_selection import train_test_split
import mglearn
import matplotlib.pyplot as plt
x, y = mglearn.datasets.make_forge()
x_train, x_test, y_train, y_test = train_test_split(x, y, random_state=0)  # 生成训练和测试集数据

from sklearn.neighbors import KNeighborsClassifier
14 clf = KNeighborsClassifier(n_neighbors=3)  # 调用K近邻分类算法
15 
16 clf.fit(x_train, y_train)  # 训练数据

print("Test set predictions:{}".format(clf.predict(x_test)))  # 预测

print("Test set accuracy:{:.2f}".format(clf.score(x_test, y_test)))

fig, axes = plt.subplots(1, 3, figsize=(10, 3))  # 使用matplotlib画图

for n_neighbors, ax in zip([1, 3, 9], axes):
    # fit 方法返回对象本身，所以我们可以将实例化和拟合放在一行代码中
    clf = KNeighborsClassifier(n_neighbors=n_neighbors).fit(x, y)
    mglearn.plots.plot_2d_separator(clf, x, fill=True, eps=0.5, ax=ax, alpha=0.4)
    mglearn.discrete_scatter(x[:, 0], x[:, 1], y, ax=ax)
    ax.set_title("{} neighbor(s)".format(n_neighbors))
    ax.set_xlabel("feature 0")
    ax.set_ylabel("feature 1")
axes[0].legend(loc=3)

结果：

总结：从图中可以看出，使用单一邻居绘制的决策边界紧跟着训练数据，随着邻居的增多，决策边界也越来越平滑，更平滑的边界对应更简单的模型，换句话说，使用更少的邻居对应更高的模型复杂度。