SVM报告-01.png

SVM报告-02.png

SVM报告-03.png

SVM报告-04.png

SVM报告-05.png

SVM报告-06.png

SVM报告-07.png

SVM报告-08.png

SVM报告-09.png

SVM报告-10.png

PR4SVM.py

import numpy as np
import matplotlib.pyplot as plt
from sklearn import svm, datasets

def main(n):
    # 导入数据集
    iris = datasets.load_iris()
    X = iris.data[:,n*2:n*2+2]  # 只取前两维特征
    y = iris.target

    h = .02  # 网格中的步长

    # 创建支持向量机实例，并拟合出数据
    C = 1.0  # SVM正则化参数
    svc = svm.SVC(kernel='linear', C=C).fit(X, y) # 线性核
    rbf_svc = svm.SVC(kernel='rbf', gamma=0.7, C=C).fit(X, y) # 径向基核
    poly_svc = svm.SVC(kernel='poly', degree=3, C=C).fit(X, y) # 多项式核
    # lin_svc = svm.LinearSVC(C=C).fit(X, y) #线性核
    # SVC : 1/2||w||^2 + C SUM xi_i  均方误差
    # LinearSVC: 1/2||[w b]||^2 + C SUM xi_i  normal hinge

    # 创建网格，以绘制图像
    x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1
    y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1
    xx, yy = np.meshgrid(np.arange(x_min, x_max, h),
                         np.arange(y_min, y_max, h))

    # 图的标题
    titles = ['SVC with linear kernel',
              # 'LinearSVC (linear kernel)',
              'SVC with RBF kernel',
              'SVC with polynomial (degree 3) kernel']


    for i, clf in enumerate((svc, rbf_svc, poly_svc)):
        # 绘出决策边界，不同的区域分配不同的颜色
        # plt.subplot(3, 1, i + 1) # 创建一个2行2列的图，并以第i个图为当前图
        # plt.subplots_adjust(wspace=0.4, hspace=0.4) # 设置子图间隔

        Z = clf.predict(np.c_[xx.ravel(), yy.ravel()])
        # 把分类结果绘制出来
        Z = Z.reshape(xx.shape) #(220, 280)
        plt.contourf(xx, yy, Z, cmap=plt.cm.Paired, alpha=0.8) #使用等高线的函数将不同的区域绘制出来

        # 将训练数据以离散点的形式绘制出来
        plt.scatter(X[:, 0], X[:, 1], c=y, cmap=plt.cm.Paired)
        plt.xlabel('Sepal length')
        plt.ylabel('Sepal width')
        plt.xlim(xx.min(), xx.max())
        plt.ylim(yy.min(), yy.max())
        plt.xticks(())
        plt.yticks(())
        plt.title(titles[i])
        plt.savefig('%d%d'%(n,i))
        plt.show()
        plt.cla()

if __name__ == '__main__':
    main(0)
    main(1)

process_data.py

import numpy as np
import csv

#%%
# To->process_data

#---usps
def get_data():
    with open('usps_all.csv','r') as f:
        reader=csv.reader(f)
        # for line in reader:
        file=np.array(list(map(np.array,reader)))
        N=len(file)
        t_N=int(0.9*N)
        data={'train':{'x':0,'y':0},'val':{'x':0,'y':0}}
        data['train']['x']=file[:t_N,:-1]
        data['train']['y']=file[:t_N,-1:]
        data['val']['x']=file[t_N:,:-1]
        data['val']['y']=file[t_N:,-1:]
    print('hello')
    return data

SVM_usps.py

import numpy as np
import matplotlib.pyplot as plt
from sklearn import svm
import process_data as pd
from sklearn.externals import joblib #jbolib模块
from os import listdir
import os
from copy import deepcopy

#读取Model
model_dir='save/'
titles = ['SVC with linear kernel',
                  'SVC with RBF kernel',
                  'SVC with polynomial (degree 3) kernel']
def main():
    # 读取数据
    data=pd.get_data()
    ## 验证集
    x,y=data['val']['x'],data['val']['y'].ravel()
    y_num=len(y)
    model_list=listdir(model_dir)

    if(model_list==[]):
        ## 训练集
        X = data['train']['x']
        Y = data['train']['y'].ravel()
        # 创建支持向量机实例，并拟合出数据
        C = 1.0  # SVM正则化参数
        print('trianing:', titles[0])
        svc = svm.SVC(kernel='linear', C=C).fit(X, Y)  # 线性核
        joblib.dump(svc, 'save/0.pkl',compress=1)

        print('trianing:', titles[1])
        rbf_svc = svm.SVC(kernel='rbf').fit(X, Y)  # 径向基核
        joblib.dump(rbf_svc, 'save/1.pkl',compress=1)

        print('trianing:', titles[2])
        poly_svc = svm.SVC(kernel='poly', degree=3, C=C).fit(X, Y)  # 多项式核
        joblib.dump(poly_svc, 'save/2.pkl',compress=1)


        exit(0)
    else:
        os.chdir(model_dir)
        svc,rbf_svc,poly_svc=list(map(joblib.load,model_list))
    # 预测

    for i, clf in enumerate((svc, rbf_svc, poly_svc)):
        y_=clf.predict(x)
        acc=sum(y_==y)/y_num
        print(titles[i]+' accuracy: ',acc)
if __name__ == '__main__':
    main()