神经网络工具箱的使用

天零ヮ冰釋

1. clear; \9 J; i, W3 O3 _( a: X
   
2. load('trainset.mat');                   %加载训练集数据
   % Q$ d3 {) w: o- a- w8 i
3. load('testset.mat');                    %加载测试集数据\" t  `# n- d9 [
   
4. p1=trainsetdata([1:179],:);
   : J7 ]! `# a5 Y8 Q; |# N: M; p
5. p2=trainsetdata([180:300],:);
   & S: U& P# ~7 e9 W8 f0 k% y\" e  }
6. p=[p1;p2]';8 h  w5 j& k2 G% v8 O' [
   
7. pr=minmax(p);
   ' h8 M, K# |1 U; n1 S  Z
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];
   3 S9 E' f5 v4 p. h# s& {% }0 m
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点$ q: v4 g+ x- K+ c! Z% k
   
10.                                             %取值不太确定，请指教
    . E1 [/ J$ M. q; e; |1 q  P+ q
11. net.trainParam.show = 10;, I0 u  m5 j  M
    
12. net.trainParam.lr = 0.05;
    6 q\" o# W* @8 G' Z1 d  z2 {7 b
13. net.trainParam.goal = 1e-10;
    $ Y% k3 C; G: p- Y9 }2 ]
14. net.trainParam.epochs = 50000;
    . G8 J/ T( K- @3 C+ e
15. net = train(net,p,goal);8 b6 q& T$ g8 o
    
16. x=testsetdata';0 a, G- @- b* |4 y& I' V2 h
    
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. / b+ Z- V. Z! a( J6 l% ?% u$ O$ q
   
4. > In obs_use at 186 H% R( n* J- Y\\" b
   
6.   In newff>create_network at 127  v1 ?9 A\\" i, k& H* o9 k7 X0 n
   
8.   In newff at 102) l/ G& r7 {5 |* E- [# H
   
10.   In test at 9
    
11. 3 Q7 ?, t\\" U\\" D% M4 y
12.           See help for NEWFF to update calls to the new argument list.$ T\\" t; O3 }3 a3 d' \( @4 F  C
    
14. 
    
15. 2 G1 @8 S; H% M# D, Q  s; P
16. ??? Error using ==> trainlm at 1097 V* Z2 a1 J: K) @% }/ m8 G
    
18. Output data size does not match net.outputs{2}.size.# d, }+ m# ]1 N
    
20. 
    
21. - j4 j* m. j, N+ q: z: q/ _
22. Error in ==> network.train at 1076 m5 n# x) k5 s. x. t8 }
    
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);
    
25.   C. Z& h( E9 Z9 N6 f1 a( Z
26. 3 c  R7 R' v' Q& p9 S
    
28. Error in ==> test at 15
    
29. % |( D5 ]2 m& D' w* ?
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <font class=\\"jammer\\">/ b+ Z- V. Z! a( J6 l% ?% u$ O$ q</font><br /> > In obs_use at 18<font class=\\"jammer\\">6 H% R( n* J- Y\\" b</font><br />   In newff>create_network at 127<font class=\\"jammer\\">  v1 ?9 A\\" i, k& H* o9 k7 X0 n</font><br />   In newff at 102<font class=\\"jammer\\">) l/ G& r7 {5 |* E- [# H</font><br />   In test at 9<br /> <span style=\\"display:none\\">3 Q7 ?, t\\" U\\" D% M4 y</span>           See help for NEWFF to update calls to the new argument list.<font class=\\"jammer\\">$ T\\" t; O3 }3 a3 d' \( @4 F  C</font><br /> <br /> <span style=\\"display:none\\">2 G1 @8 S; H% M# D, Q  s; P</span> ??? Error using ==> trainlm at 109<font class=\\"jammer\\">7 V* Z2 a1 J: K) @% }/ m8 G</font><br /> Output data size does not match net.outputs{2}.size.<font class=\\"jammer\\"># d, }+ m# ]1 N</font><br /> <br /> <span style=\\"display:none\\">- j4 j* m. j, N+ q: z: q/ _</span> Error in ==> network.train at 107<font class=\\"jammer\\">6 m5 n# x) k5 s. x. t8 }</font><br /> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<br /> <span style=\\"display:none\\">  C. Z& h( E9 Z9 N6 f1 a( Z</span> <font class=\\"jammer\\">3 c  R7 R' v' Q& p9 S</font><br /> Error in ==> test at 15<br /> <span style=\\"display:none\\">% |( D5 ]2 m& D' w* ?</span> net = train(net,p,goal);

jiajinshan110

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