神经网络工具箱的使用

天零ヮ冰釋

1. clear# Y( d; N5 n, S9 n' J
   
2. load('trainset.mat');                   %加载训练集数据$ E$ d0 M\" d% ?\" p& Y& i' G
   
3. load('testset.mat');                    %加载测试集数据
   : _1 L  {5 t( J  u; A
4. p1=trainsetdata([1:179],:);
   3 \, _$ w* v3 G
5. p2=trainsetdata([180:300],:);3 s5 P, i$ U* m' A
   
6. p=[p1;p2]';
   & a\" B$ s+ ^8 n$ w
7. pr=minmax(p);; v/ S! X' G9 Q7 t# q' W
   
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];
   2 A) y' [) @% a7 K  @! S( ~7 J
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点3 e; n# i% `* {. w* y9 A\" [( F3 E. D
   
10.                                             %取值不太确定，请指教( f& W9 E5 L0 }% `7 F
    
11. net.trainParam.show = 10;
    6 C9 X1 t1 d+ ^
12. net.trainParam.lr = 0.05;; @. e2 D8 s* D6 i& h7 z2 T
    
13. net.trainParam.goal = 1e-10;
    9 z' \/ f  D/ G* G# f: U
14. net.trainParam.epochs = 50000;4 {\" ^% w0 z. U6 j% i3 q$ |6 c9 l
    
15. net = train(net,p,goal);
    + R( r9 U- b2 r\" y& W
16. x=testsetdata';- X, s* t0 c+ ^8 h0 _% _( i
    
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. 5 ?4 h1 O2 s/ `$ _
   
4. > In obs_use at 18
   
5. ! d* T, \6 B7 w, N
6.   In newff>create_network at 127! M. S* Y) P( ^
   
8.   In newff at 102
   
9. $ w8 v! X, D# P8 ~, c. Z
10.   In test at 9+ d% E7 Y* g# @; D1 U; Y
    
12.           See help for NEWFF to update calls to the new argument list.
    
13. / G. N- M9 O\\" Y; s9 g
14. 
    
15. 2 s* U9 d+ q! R4 o/ B  a
16. ??? Error using ==> trainlm at 109
    
17. ; }! y& U& e& b. T. p7 {
18. Output data size does not match net.outputs{2}.size.7 k7 L+ T: G4 G% ?& v
    
20. 
    
21. 1 f9 F! j9 d/ R; `* K8 l) Z
22. Error in ==> network.train at 1079 g* x1 d' F, T$ q9 B
    
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);: I6 o. G& L5 p! q0 K( |
    
26. * o1 m9 G* B0 \3 J; t7 A9 {
    
28. Error in ==> test at 15. ?6 W8 |  M8 k( n. b+ _2 W
    
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <font class=\\"jammer\\">5 ?4 h1 O2 s/ `$ _</font><br /> > In obs_use at 18<br /> <span style=\\"display:none\\">! d* T, \6 B7 w, N</span>   In newff>create_network at 127<font class=\\"jammer\\">! M. S* Y) P( ^</font><br />   In newff at 102<br /> <span style=\\"display:none\\">$ w8 v! X, D# P8 ~, c. Z</span>   In test at 9<font class=\\"jammer\\">+ d% E7 Y* g# @; D1 U; Y</font><br />           See help for NEWFF to update calls to the new argument list.<br /> <span style=\\"display:none\\">/ G. N- M9 O\\" Y; s9 g</span> <br /> <span style=\\"display:none\\">2 s* U9 d+ q! R4 o/ B  a</span> ??? Error using ==> trainlm at 109<br /> <span style=\\"display:none\\">; }! y& U& e& b. T. p7 {</span> Output data size does not match net.outputs{2}.size.<font class=\\"jammer\\">7 k7 L+ T: G4 G% ?& v</font><br /> <br /> <span style=\\"display:none\\">1 f9 F! j9 d/ R; `* K8 l) Z</span> Error in ==> network.train at 107<font class=\\"jammer\\">9 g* x1 d' F, T$ q9 B</font><br /> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<font class=\\"jammer\\">: I6 o. G& L5 p! q0 K( |</font><br /> <font class=\\"jammer\\">* o1 m9 G* B0 \3 J; t7 A9 {</font><br /> Error in ==> test at 15<font class=\\"jammer\\">. ?6 W8 |  M8 k( n. b+ _2 W</font><br /> net = train(net,p,goal);

jiajinshan110

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