1 支持向量机
(简称SVM)是一个功能强大并且全面的机器学习模型,它能够执行线性或非线性分类、回归,甚至是异常值检测任
务。它是机器学习领域最受欢迎的模型之一,任何对机器学习感兴趣的人都应该在工具箱中配备一个。SVM特别适用于中小型复杂数据集的分类。
2 案例
# -*- coding: utf-8 -*-
import scipy.io as scio
wineData = scio.loadmat('D:\\DATA\\pycase\\number2\\5.5\\Data.mat')
wineData['categories']
fData = wineData['wine']
tData = wineData['wine_labels'].reshape(-1)
from sklearn import svm
from sklearn.model_selection import cross_val_score
# 比较三种方法的评分,选择最高
svmModel = svm.SVC()
# K折交叉验证
cross_val_score(
svmModel,
fData,
tData,
cv=3
)
svmModel = svm.NuSVC()
cross_val_score(
svmModel,
fData, tData, cv=3
)
svmModel = svm.LinearSVC()
cross_val_score(
svmModel,
fData, tData, cv=3
)
svmModel = svm.LinearSVC()
svmModel.fit(fData, tData)
svmModel.score(fData, tData)