神经网络工具箱的使用

天零ヮ冰釋

1. clear' t\" w- ?$ ]\" [4 ^3 @6 y9 l
   
2. load('trainset.mat');                   %加载训练集数据$ ]- [9 w. k. a0 d# U7 r
   
3. load('testset.mat');                    %加载测试集数据% g4 h  |\" I* r- A/ {2 Z. Q0 l
   
4. p1=trainsetdata([1:179],:);. }5 ]' r3 J, c7 B/ @6 q3 x
   
5. p2=trainsetdata([180:300],:);, b* {* {1 _4 y  Q) w0 r
   
6. p=[p1;p2]';
   * y+ P# _' x- G$ G* g( x
7. pr=minmax(p);- z  V0 n8 a3 X+ P( o% F
   
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];+ b! I& t- x% k3 S
   
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点
   8 w0 p& U7 x: n2 ?2 P1 _
10.                                             %取值不太确定，请指教
    % J  W$ U/ X* E- G, o
11. net.trainParam.show = 10;5 z; Q# ~1 w: ]) j. |! w0 R
    
12. net.trainParam.lr = 0.05;9 C( `* W7 O\" B
    
13. net.trainParam.goal = 1e-10;
    \" A, t7 T# o2 `\" W. e: i
14. net.trainParam.epochs = 50000;/ _9 N2 q0 H5 p. ]- }
    
15. net = train(net,p,goal);% g\" e' F# e8 u& D6 @
    
16. x=testsetdata';; ]7 |  _! ]3 `! 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. 2 q5 U; b0 ^, X3 |6 q
4. > In obs_use at 18
   
5. . x. w% t1 R, E2 u
6.   In newff>create_network at 127- e  B3 D. x' ]) D6 ^* k# F
   
8.   In newff at 1027 d0 h/ c) `+ z, V* U
   
10.   In test at 9
    
11. ! D  t) F* x  @2 G
12.           See help for NEWFF to update calls to the new argument list.
    
13. 3 O\\" k' {0 u- i# I: q) D
14. 8 c' u& z2 O$ M! K( Z
    
16. ??? Error using ==> trainlm at 109
    
17. . p- @, v' Y$ {# b& v
18. Output data size does not match net.outputs{2}.size.
    
19. * _2 i) u0 |8 Y8 v5 D1 z9 [
20. & }0 \, E$ T. s3 z7 E' p\\" n
    
22. Error in ==> network.train at 107
    
23. . {7 Q  W) H0 f
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);! m$ C  U( n& T% {' r
    
26. 
    
27. + l# q9 x+ c2 b6 p8 v: K
28. Error in ==> test at 15) L* |) _0 A, z) ^2 w' R
    
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <br /> <span style=\\"display:none\\">2 q5 U; b0 ^, X3 |6 q</span> > In obs_use at 18<br /> <span style=\\"display:none\\">. x. w% t1 R, E2 u</span>   In newff>create_network at 127<font class=\\"jammer\\">- e  B3 D. x' ]) D6 ^* k# F</font><br />   In newff at 102<font class=\\"jammer\\">7 d0 h/ c) `+ z, V* U</font><br />   In test at 9<br /> <span style=\\"display:none\\">! D  t) F* x  @2 G</span>           See help for NEWFF to update calls to the new argument list.<br /> <span style=\\"display:none\\">3 O\\" k' {0 u- i# I: q) D</span> <font class=\\"jammer\\">8 c' u& z2 O$ M! K( Z</font><br /> ??? Error using ==> trainlm at 109<br /> <span style=\\"display:none\\">. p- @, v' Y$ {# b& v</span> Output data size does not match net.outputs{2}.size.<br /> <span style=\\"display:none\\">* _2 i) u0 |8 Y8 v5 D1 z9 [</span> <font class=\\"jammer\\">& }0 \, E$ T. s3 z7 E' p\\" n</font><br /> Error in ==> network.train at 107<br /> <span style=\\"display:none\\">. {7 Q  W) H0 f</span> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<font class=\\"jammer\\">! m$ C  U( n& T% {' r</font><br /> <br /> <span style=\\"display:none\\">+ l# q9 x+ c2 b6 p8 v: K</span> Error in ==> test at 15<font class=\\"jammer\\">) L* |) _0 A, z) ^2 w' R</font><br /> net = train(net,p,goal);

jiajinshan110

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