神经网络工具箱的使用

天零ヮ冰釋

1. clear2 u; m& B5 ^, z, v! J4 a& {9 S
   
2. load('trainset.mat');                   %加载训练集数据
   ( C: q, a5 L\" g0 h. Y7 a
3. load('testset.mat');                    %加载测试集数据
   0 B* j# B4 q% g2 L
4. p1=trainsetdata([1:179],:);  {# B' {1 Q3 N! G% U* `
   
5. p2=trainsetdata([180:300],:);1 Y1 q: V' }6 y+ n
   
6. p=[p1;p2]';
   ; t& G: e; \7 Y6 t
7. pr=minmax(p);
   + W( M- h) k- T. H6 c9 ^& [( u
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];$ M: m8 i$ S8 K
   
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点/ _, u$ Y- x- k) w. a
   
10.                                             %取值不太确定，请指教
    ; O* ?: c) w' A
11. net.trainParam.show = 10;
    * q2 c& D* v9 [( H. P: x/ s& ^
12. net.trainParam.lr = 0.05;3 L% D9 h: x8 X3 H% j
    
13. net.trainParam.goal = 1e-10;
    5 n( o9 i6 n5 b( A( b% y% ~
14. net.trainParam.epochs = 50000;5 h* q8 y* K+ D; ]. P& W
    
15. net = train(net,p,goal);
    8 I! f\" v+ H\" O* ?: [) M
16. x=testsetdata';
    2 i9 s) Y+ `5 d  e) f4 s1 L/ E
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. 7 u2 F/ {3 s# w2 ^; X
   
4. > In obs_use at 18
   
5. * f5 a\\" @7 i& x# A# i: j
6.   In newff>create_network at 127
   
7. - o3 ^9 l2 a& t3 m
8.   In newff at 102
   
9. ! }& O% _8 q7 V\\" U/ l
10.   In test at 95 v$ i, A) ?5 U8 r4 x& r8 d0 ^
    
12.           See help for NEWFF to update calls to the new argument list.
    
13. , M  {7 i) Z: E) t9 Y& G
14. 
    
15. 2 q/ Z  g0 d! \% \5 g
16. ??? Error using ==> trainlm at 109
    
17. # r! ?3 A\\" |: p- i, R$ ]' s' a7 i, t
18. Output data size does not match net.outputs{2}.size.
    
19. . G- l4 s6 {5 H4 N\\" w6 b$ |
20. 
    
21. + S0 s8 r7 m/ U( H
22. Error in ==> network.train at 107# |8 M% ?0 d' i  D3 e
    
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);
    
25. 5 m! k, f, \  I) k0 `* m2 I- I
26. . G+ s4 m& v) j; i( E\\" H0 \/ w
    
28. Error in ==> test at 15
    
29. % [! k% H# D9 l. I3 P/ x* [( ~
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <font class=\\"jammer\\">7 u2 F/ {3 s# w2 ^; X</font><br /> > In obs_use at 18<br /> <span style=\\"display:none\\">* f5 a\\" @7 i& x# A# i: j</span>   In newff>create_network at 127<br /> <span style=\\"display:none\\">- o3 ^9 l2 a& t3 m</span>   In newff at 102<br /> <span style=\\"display:none\\">! }& O% _8 q7 V\\" U/ l</span>   In test at 9<font class=\\"jammer\\">5 v$ i, A) ?5 U8 r4 x& r8 d0 ^</font><br />           See help for NEWFF to update calls to the new argument list.<br /> <span style=\\"display:none\\">, M  {7 i) Z: E) t9 Y& G</span> <br /> <span style=\\"display:none\\">2 q/ Z  g0 d! \% \5 g</span> ??? Error using ==> trainlm at 109<br /> <span style=\\"display:none\\"># r! ?3 A\\" |: p- i, R$ ]' s' a7 i, t</span> Output data size does not match net.outputs{2}.size.<br /> <span style=\\"display:none\\">. G- l4 s6 {5 H4 N\\" w6 b$ |</span> <br /> <span style=\\"display:none\\">+ S0 s8 r7 m/ U( H</span> Error in ==> network.train at 107<font class=\\"jammer\\"># |8 M% ?0 d' i  D3 e</font><br /> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<br /> <span style=\\"display:none\\">5 m! k, f, \  I) k0 `* m2 I- I</span> <font class=\\"jammer\\">. G+ s4 m& v) j; i( E\\" H0 \/ w</font><br /> Error in ==> test at 15<br /> <span style=\\"display:none\\">% [! k% H# D9 l. I3 P/ x* [( ~</span> net = train(net,p,goal);

jiajinshan110

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