p=[0.3 0.923;1.0 0.934;4 0.924;20 0.916;20 0.927;45 0.9555;12 0.952;23 0.9575;6.5 0.9585;1.6 0.948;0.5 0.948;0.5 0.9445;1.65 0.943;0.375 0.9455;2.0 0.919;20 0.921;27 0.9515;1.6 0.943;1.6 0.947]';
t=[15.17 1.7;19.17 1.4;13.10 1.4;8.83 1.5;15.10 1.1;32 2.0;30 5.0;32 3.0;33 4.5;30 8;30 20;26 30;24 10;24 12;12.55 1.6;11.58 1.4;29 2.8;26 10;29 10]';
%歸一化
u=t;
for i=1:2
   P(i,:)=(p(i,:)-min(p(i,:)))/(max(p(i,:))-min(p(i,:)));
end
for i=1:2
    T(i,:)=(t(i,:)-min(t(i,:)))/(max(t(i,:))-min(t(i,:)));
end
%測試樣本
PP=[P(:,1) P(:,2) P(:,3) P(:,4) P(:,5) P(:,6) P(:,7) P(:,8) P(:,9) P(:,10) P(:,11) P(:,12) P(:,13) P(:,14)];
TT=[T(:,1) T(:,2) T(:,3) T(:,4) T(:,5) T(:,6) T(:,7) T(:,8) T(:,9) T(:,10) T(:,11) T(:,12) T(:,13) T(:,14)];
P_test=[P(:,15) P(:,16) P(:,17) P(:,18) P(:,19)];
T_test=[T(:,15) T(:,16) T(:,17) T(:,18) T(:,19)];
net=newff(minmax(PP),[5,2],{'tansig','logsig'},'trainlm');
net.trainParam.epochs=1000;
net.trainParam.goal=0.01;
LP.lr=0.1;
net=train(net,PP,TT);
out=sim(net,P_test);
%反歸一化
for i=1:2
   predict(i,1)=out(i,1)* (max(u(i,:))-min(u(i,:)))+ min(u(i,:));
   predict(i,2)=out(i,2)* (max(u(i,:))-min(u(i,:)))+ min(u(i,:));
   predict(i,3)=out(i,3)* (max(u(i,:))-min(u(i,:)))+ min(u(i,:));
   predict(i,4)=out(i,4)* (max(u(i,:))-min(u(i,:)))+ min(u(i,:));
   predict(i,5)=out(i,5)* (max(u(i,:))-min(u(i,:)))+ min(u(i,:));

end
predict
X=1:5;
figure(1);
plot(X,predict(1,:),'r*',X,t(1,15:19),'bo')

xlabel('你自己填');
ylabel('你自己填');

text(4,predict(1,4),'predicted data');