normalization.zip_MATLAB归一化_Normalization_matlab 归一化_数据处理_数据归一化

function normal = normalization(y,kind) load 'powerplant.dat'; y=[powerplant]; testlabel=1:200; testlabel=testlabel'; q1=rand(1200,1)*2; q2=zeros(2400,1); q2(1:200,1)=y(:,1); q2(200:400,1)=y(:,2); q2(400:600,1)=y(:,3); q2(600:800,1)=y(:,4); q2(800:1000,1)=y(:,5); [m,n]=size(q2); trainin=[q1(1:1000,1),q2(2:1000,1)]; trainout=q2(1600:1800,1); testin=[q1(1000:1200,1),q2(1800:2000,1)]; testout=q2(2000:2200,1); [X,Y] = meshgrid(-10:10,-10:10); [m,n] = size(X); cg = zeros(m,n); bestc = 0; bestg = 0; v=3; mse = 10^10; basenum = 2; for i = 1:m for j = 1:n cmd = ['-v ',num2str(v),' -c ',num2str( basenum^X(i,j) ),' -g ',num2str( basenum^Y(i,j) ),' -s 3']; cg(i,j) = svmtrain(trainout,trainin, cmd); if cg(i,j) < mse mse = cg(i,j); bestc = basenum^X(i,j); bestg = basenum^Y(i,j); end if ( cg(i,j) == mse && bestc > basenum^X(i,j) ) mse = cg(i,j); bestc = basenum^X(i,j); bestg = basenum^Y(i,j); end end end fprintf('(best c=%g, g=%g, mse=%g)\n',bestc, bestg, mse); cmd = ['-c ', num2str(bestc), ' -g ', num2str(bestg) , ' -s 3 -e 0.001']; model = svmtrain(trainout,trainin, cmd); [predict,accuracy] = svmpredict(testlabel,testin,model); for i=1:200 mse(i)=predict(i)-testout(i); end hold on; plot(mse,'-b.'); plot(testout,'-ko'); grid on; plot(predict,'-g*','LineWidth',2); grid on; legend('mse','testout','predict');