matlab30个案例分析案例13-SVM神经网络中的参数优化.docx

%% SVM神经网络中的参数优化---如何更好的提升分类器的性能%% 清空环境变量function chapter13_GAclose all;clear;clc;format compact;%% 数据提取% 载入测试数据wine,其中包含的数据为classnumber = 3,wine:178*13的矩阵,wine_labes:178*1的列向量load chapter13_wine.mat;% 画出测试数据的box可视化图figure;boxplot(wine,orientation,horizontal,labels,categories);title(wine数据的box可视化图,FontSize,12);xlabel(属性值,FontSize,12);grid on;% 画出测试数据的分维可视化图figuresubplot(3,5,1);hold onfor run = 1:178plot(run,wine_labels(run),*);endxlabel(样本,FontSize,10);ylabel(类别标签,FontSize,10);title(class,FontSize,10);for run = 2:14subplot(3,5,run);hold on;str = [attrib ,num2str(run-1)];fori = 1:178plot(i,wine(i,run-1),*);endxlabel(样本,FontSize,10);ylabel(属性值,FontSize,10);title(str,FontSize,10);end% 选定训练集和测试集% 将第一类的1-30,第二类的60-95,第三类的131-153做为训练集train_wine = [wine(1:30,:);wine(60:95,:);wine(131:153,:)];% 相应的训练集的标签也要分离出来train_wine_labels = [wine_labels(1:30);wine_labels(60:95);wine_labels(131:153)];% 将第一类的31-59,第二类的96-130,第三类的154-178做为测试集test_wine = [wine(31:59,:);wine(96:130,:);wine(154:178,:)];% 相应的测试集的标签也要分离出来test_wine_labels = [wine_labels(31:59);wine_labels(96:130);wine_labels(154:178)];%% 数据预处理% 数据预处理,将训练集和测试集归一化到[0,1]区间[mtrain,ntrain] = size(train_wine);[mtest,ntest] = size(test_wine);dataset = [train_wine;test_wine];% mapminmax为MATLAB自带的归一化函数[dataset_scale,ps] = mapminmax(dataset,0,1);dataset_scale = dataset_scale;train_wine = dataset_scale(1:mtrain,:);test_wine = dataset_scale( (mtrain+1):(mtrain+mtest),: );%% 选择GA最佳的SVM参数cg% GA的参数选项初始化ga_option.maxgen = 200;ga_option.sizepop = 20; ga_option.cbound = [0,100];ga_option.gbound = [0,100];ga_option.v = 5;ga_option.ggap = 0.9;[bestacc,bestc,bestg] = gaSVMcgForClass(train_wine_labels,train_wine,ga_option);% 打印选择结果disp(打印选择结果);str = sprintf( Best Cross Validation Accuracy = %g%% Best c = %g Best g = %g,bestacc,bestc,bestg);disp(str);%% 利用最佳的参数进行SVM网络训练cmd = [-c ,num2str(bestc), -g ,num2str(bestg)];model = svmtrain(train_wine_labels,train_wine,cmd);%% SVM网络预测[predict_label,accuracy] = svmpredict(test_wine_labels,test_wine,model);% 打印