神经网络工具箱的使用

天零ヮ冰釋

1. clear# m3 T4 G' S( W& j) q, {5 m7 U/ ?0 L
   
2. load('trainset.mat');                   %加载训练集数据
   . s. x+ W. n' ~
3. load('testset.mat');                    %加载测试集数据1 g2 K' @# }: j
   
4. p1=trainsetdata([1:179],:);
   ) ^: R5 }& h- O( r6 q# C( f
5. p2=trainsetdata([180:300],:);; s4 d% ?$ P. g! F4 @, N1 z' p
   
6. p=[p1;p2]';8 O5 r4 H; {' O) L- [\" v
   
7. pr=minmax(p);* ^( M' q\" x$ z* j0 e1 S; }
   
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];
   & n/ B0 v; K3 J2 Q
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点
   : A4 J: F( E2 U1 @\" V
10.                                             %取值不太确定，请指教+ ~6 B. R: Z4 e\" f\" y) {
    
11. net.trainParam.show = 10;, b/ @' j1 @6 j, b
    
12. net.trainParam.lr = 0.05;1 k6 H. }$ @) G8 R7 ^\" t
    
13. net.trainParam.goal = 1e-10;, X8 O) B1 N5 Q* \3 N6 B0 k
    
14. net.trainParam.epochs = 50000;
    - c, l9 G* j6 C2 z
15. net = train(net,p,goal);6 `/ W$ l: f% x
    
16. x=testsetdata';
    ! l% O( U  j; Z1 Z* _* R! z; V
17. y0=sim(net,p)

复制代码

下面是运行结果，请各位高手帮忙改改啊，主要是节点数那块儿不知道该怎么取值

2. Warning: NEWFF used in an obsolete way.
   
3. ) J\\" B9 ~' D' y- ?
4. > In obs_use at 18
   
5. % a2 s6 V! ^! M+ o
6.   In newff>create_network at 127$ Y% B# M; X( G3 d7 ^
   
8.   In newff at 102
   
9. 5 x2 ^; t) F5 b1 v5 j# O
10.   In test at 9
    
11. 9 ?( o) v( ?. Q+ X. E
12.           See help for NEWFF to update calls to the new argument list.
    
13. 7 A; Z) P. i$ [$ j) a/ D2 R. x( ]
14. 2 D' [# w1 J6 |( r. L
    
16. ??? Error using ==> trainlm at 109
    
17. 4 _4 a9 [' K% L* ^0 ^
18. Output data size does not match net.outputs{2}.size.( }) t3 \5 C3 R
    
20. 0 a' m8 w# H, m  k0 A! r5 H& J
    
22. Error in ==> network.train at 107) O+ q1 ?. B# I! T' s
    
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);( P# E/ [+ d  y' s% }\\" }
    
26. 9 ]  E9 p: g( \- V5 C8 L. Z9 C- N
    
28. Error in ==> test at 15
    
29. & K* N! M% T; \
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <br /> <span style=\\"display:none\\">) J\\" B9 ~' D' y- ?</span> > In obs_use at 18<br /> <span style=\\"display:none\\">% a2 s6 V! ^! M+ o</span>   In newff>create_network at 127<font class=\\"jammer\\">$ Y% B# M; X( G3 d7 ^</font><br />   In newff at 102<br /> <span style=\\"display:none\\">5 x2 ^; t) F5 b1 v5 j# O</span>   In test at 9<br /> <span style=\\"display:none\\">9 ?( o) v( ?. Q+ X. E</span>           See help for NEWFF to update calls to the new argument list.<br /> <span style=\\"display:none\\">7 A; Z) P. i$ [$ j) a/ D2 R. x( ]</span> <font class=\\"jammer\\">2 D' [# w1 J6 |( r. L</font><br /> ??? Error using ==> trainlm at 109<br /> <span style=\\"display:none\\">4 _4 a9 [' K% L* ^0 ^</span> Output data size does not match net.outputs{2}.size.<font class=\\"jammer\\">( }) t3 \5 C3 R</font><br /> <font class=\\"jammer\\">0 a' m8 w# H, m  k0 A! r5 H& J</font><br /> Error in ==> network.train at 107<font class=\\"jammer\\">) O+ q1 ?. B# I! T' s</font><br /> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<font class=\\"jammer\\">( P# E/ [+ d  y' s% }\\" }</font><br /> <font class=\\"jammer\\">9 ]  E9 p: g( \- V5 C8 L. Z9 C- N</font><br /> Error in ==> test at 15<br /> <span style=\\"display:none\\">& K* N! M% T; \</span> net = train(net,p,goal);

jiajinshan110

..................