Doubt regards the Back Propagation Network training

1 visualización (últimos 30 días)
ajith
ajith el 18 de Jul. de 2013
sir my doubt regards the coding process for training the BPN is correct or not sir, would you been specified the error would i did thank you sir and also specified the newer version of train for BPN
input1=q(1,1:419); output1(1:419)=0;
input2=q(2,1:419); output2(1:419)=0;
input3=q(3,1:419); output3(3:419)=0;
input4=q(4,1:419); output4(4:419)=0;
input5=q(5,1:419); output5(5:419)=0;
input6=q(6,1:419); output6(6:419)=0;
input7=q(7,1:419); output7(7:419)=1;
input8=q(8,1:419); output8(8:419)=1;
input9=q(9,1:419); output9(9:419)=1;
input10=q(10,1:419); output10(10:419)=1;
input = [input1 input2 input3 input4 input5 input6 input7 input8 input9 input10]; output = [output1 output2 output3 output4 output5 output6 output7 output8 output9 output10];
val=[min(input(1,:)) max(input(1,:))];
nsch=newff(val,[100 1],{'tansig' 'purelin'},'trainlm');
nsch.trainparam.show=2;
nsch.trainparam.lr=0.01;
nsch.trainparam.mc=0.9;
nsch.trainparam.epochs=100;
nsch.trainparam.goal=1e-3;
%net.trainParam.time inf
nsch=train(nsch,input,output);
res=sim(nsch,input);

Respuestas (0)

Categorías

Más información sobre Parallel and Cloud en Help Center y File Exchange.

Etiquetas

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by