How can I ameliorate the Neural network implementation?

Question

Sa rah 2016-1-2

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/262452-how-can-i-ameliorate-the-neural-network-implementation

评论： Greg Heath 2016-1-5

I used these code lines to built a Neural Network classifier in a face recognition project:

 net = patternnet(10);
 [net,tr] = train(net,feaVectors,labels); % feaVectors=4800*90 and labels=15*90
testY = net(mat_test); % mat_test=4800*75
[c,cm] = confusion(labels_test,testY);

The problem is that I don't get a satisfying result, the true recognition rate is around 50% or less, and that's too low. I don't get why because I think the implementation is correct !! In fact, I use the same inputs with SVM classifier and I get very satisfying results. can you help me in this ?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Greg Heath 2016-1-3

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/262452-how-can-i-ameliorate-the-neural-network-implementation#answer_204911

编辑：Walter Roberson 2016-1-3

在 MATLAB Online 中打开

Try not to have have more unknown weights than training equations (OVERFITTING)because there is no guarantee that you will get good performance on nontraining data.
Rank multiple designs by the performance on validation data
Obtain unbiased performance estiamtes using the performance on test data.
It is preferrable to have more training equations than unknown weights. Therefore it is desirable to use as few hidden nodes as possible.
There is a default 0.7/0.15/0.15 RANDOM Ntrn/Nval/Ntst data division
The default initial weights are RANDOM
Given a sufficient number of hidden nodes, you may have to train 10 or 20 nets to obtain one that yields a good set of initial weights.

 [ I N ] = size(input)   % [ 4800 90 ]
 [ O N ] = size(target)  % [ 15 90 ]
 Ntst = round(0.15*N)    % 14
 Nval = Ntst             % 14
 Ntrn = N - Nval-Ntst    % 62
 Ntrneq = Ntrn*O         % 930  training equations
 H = 10                  % default # of training equations
 Nw = (I+1)*H+(H+1)*O = 48175  % Unknown weights

You have more than 50 times as many weights than you have equations (SEVERE OVERFITTING). You should try to use image feature extraction to drastically reduce the number of inputs.

Also, use the training record tr to separate training, validation and testing performance.

Hope this helps.

Thank you for formally accepting my answer

Greg

PS: I have zillions of patternnet posts in the NEWSGROUP and ANSWERS. For initial searches try

patternnet tutorial

and

patternnet greg

2 个评论
显示无隐藏无

Sa rah 2016-1-3

编辑：Sa rah 2016-1-3

thank you for your answer, but can you please tell me what algorithm do you suggest to use for feature extraction so that the inputs have less dimensions? Do you agree on PCA for example?