神经网络工具箱的使用

天零ヮ冰釋

1. clear/ |' X7 ^\" ]+ c& ?# Q4 U: J
   
2. load('trainset.mat');                   %加载训练集数据7 M9 @3 E( \  U0 {
   
3. load('testset.mat');                    %加载测试集数据
   , V: |) }3 O+ ?
4. p1=trainsetdata([1:179],:);
   . F8 y. D# C+ o) a$ D$ \
5. p2=trainsetdata([180:300],:);
   5 X  b) `\" t$ d* [
6. p=[p1;p2]';
   . ?; c& I' _1 l3 @# i' D# a: p
7. pr=minmax(p);% s. i- E1 Q% T) c
   
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];
   . v5 A1 {  |2 n6 D% @( D
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点
   , p9 @  a  h2 ~% q- B6 X
10.                                             %取值不太确定，请指教
      P0 k9 Z# N$ d4 \
11. net.trainParam.show = 10;
    ! V+ t; Q  F' g7 u4 d
12. net.trainParam.lr = 0.05;
    % _! A: G# N/ N3 a0 |
13. net.trainParam.goal = 1e-10;) p5 I7 ]' S8 k$ P' f\" u# H
    
14. net.trainParam.epochs = 50000;
    / s  \7 C1 J7 S( ^4 l- E' W
15. net = train(net,p,goal);) h/ e+ B7 W7 R  J
    
16. x=testsetdata';
    $ z2 d. c9 X8 x& f& _
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. ; C; A8 S  a\\" n& ?
4. > In obs_use at 18$ i0 q0 e  K* H2 l+ v9 T/ v
   
6.   In newff>create_network at 127  u. S& R  E* `* G# j+ H, T# |
   
8.   In newff at 102
   
9. / {\\" g  t* f: R
10.   In test at 9/ r0 e$ {, ]6 L0 W: ?
    
12.           See help for NEWFF to update calls to the new argument list.
    
13. - |\\" k! }6 \) M0 @) \2 s, \
14. 
    
15. - F6 L' H& v5 n
16. ??? Error using ==> trainlm at 109
    
17. 8 Y2 E9 t0 v1 I5 B$ A+ @( E/ y) b
18. Output data size does not match net.outputs{2}.size.9 s1 v- I0 Z) a
    
20. 
    
21. 8 d. W% T4 D1 {5 R! ^\\" X
22. Error in ==> network.train at 107; S7 d6 N2 S  f
    
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);5 ]; ~% f5 w/ _2 L
    
26. ) ]4 ~9 T7 M+ r1 t5 ]
    
28. Error in ==> test at 15$ X$ I0 \/ y- q
    
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <br /> <span style=\\"display:none\\">; C; A8 S  a\\" n& ?</span> > In obs_use at 18<font class=\\"jammer\\">$ i0 q0 e  K* H2 l+ v9 T/ v</font><br />   In newff>create_network at 127<font class=\\"jammer\\">  u. S& R  E* `* G# j+ H, T# |</font><br />   In newff at 102<br /> <span style=\\"display:none\\">/ {\\" g  t* f: R</span>   In test at 9<font class=\\"jammer\\">/ r0 e$ {, ]6 L0 W: ?</font><br />           See help for NEWFF to update calls to the new argument list.<br /> <span style=\\"display:none\\">- |\\" k! }6 \) M0 @) \2 s, \</span> <br /> <span style=\\"display:none\\">- F6 L' H& v5 n</span> ??? Error using ==> trainlm at 109<br /> <span style=\\"display:none\\">8 Y2 E9 t0 v1 I5 B$ A+ @( E/ y) b</span> Output data size does not match net.outputs{2}.size.<font class=\\"jammer\\">9 s1 v- I0 Z) a</font><br /> <br /> <span style=\\"display:none\\">8 d. W% T4 D1 {5 R! ^\\" X</span> Error in ==> network.train at 107<font class=\\"jammer\\">; S7 d6 N2 S  f</font><br /> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<font class=\\"jammer\\">5 ]; ~% f5 w/ _2 L</font><br /> <font class=\\"jammer\\">) ]4 ~9 T7 M+ r1 t5 ]</font><br /> Error in ==> test at 15<font class=\\"jammer\\">$ X$ I0 \/ y- q</font><br /> net = train(net,p,goal);

jiajinshan110

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