bp神經(jīng)網(wǎng)絡(luò)MATLAB程序

bp神經(jīng)網(wǎng)絡(luò)MATLAB程序bp神經(jīng)網(wǎng)絡(luò)MATLAB程序bp神經(jīng)網(wǎng)絡(luò)MATLAB程序bp神經(jīng)網(wǎng)絡(luò)MATLAB程序編制僅供參考審核批準(zhǔn)生效日期地址：電話(huà)：傳真:郵編:%清空環(huán)境變量clcclear%讀取數(shù)據(jù)a=xlsread('F:\4.數(shù)據(jù)挖掘講義—馬景義\數(shù)據(jù)和程序\matlab\');b=xlsread('F:\4.數(shù)據(jù)挖掘講義—馬景義\數(shù)據(jù)和程序\matlab\');c=xlsread('F:\4.數(shù)據(jù)挖掘講義—馬景義\數(shù)據(jù)和程序\matlab\');savebsavealoaddata1loaddata2%節(jié)點(diǎn)個(gè)數(shù)inputnum=42;hiddennum=8;outputnum=1;%訓(xùn)練數(shù)據(jù)和預(yù)測(cè)數(shù)據(jù)input_train=a(:,1:90);input_test=a(:,91:137);output_train=b(:,1:90);output_test=b(:,91:137);%將訓(xùn)練樣本輸入輸出數(shù)據(jù)歸一化[inputn,inputps]=mapminmax(input_train);[outputn,outputps]=mapminmax(output_train);%構(gòu)建BP神經(jīng)網(wǎng)絡(luò)net=newff(inputn,outputn,[20,12],{'tansig','purelin'},'traingdx');神經(jīng)網(wǎng)絡(luò)訓(xùn)練net=train(net,inputn,outputn);%測(cè)試樣本歸一化inputn_test=mapminmax('apply',input_test,inputps);%BP神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)an=sim(net,inputn_test);%%網(wǎng)絡(luò)得到數(shù)據(jù)反歸一化BPoutput=mapminmax('reverse',an,outputps);d=BPoutput>cfigure(1)%plot(d,':og');scatter(1:(137-90),d,'rx');holdon;%plot(output_test,'-*');scatter(1:(137-90),output_test,'o');legend('預(yù)測(cè)輸出','期望輸出','fontsize',12);title('BP網(wǎng)絡(luò)預(yù)測(cè)輸出','fontsize',12);xlabel('樣本','fontsize',12);xlabel('優(yōu)化前輸出的誤差','fontsize',12);figure(2)error=d-output_test;plot(1:(137-90),error);xlabel('樣本','fontsize',12)