MATLAB: Neural Network Design.

neural network

Please I am designing a simple neural network with the following dimensions: 35 input vectors 10 output neurons 2 hidden layers with 10 neurons each please I need a valid code to execute this design using the feed-forward back propagation algorithm;especially how to set my targets. Also I need parameters to check the network performance especially error rate. I have already done my preprocessing and features extractions. Thank you very much and I look forward to constructive ideas

Best Answer

Insufficient Information.

What versions of Matlab & NNToolbox?

What kind of physical problem are you trying to solve?

Regression or Classification?

What are your inputs and outputs?

Number of input/output training pairs: Ntrn = 35

Input vector dimensionality: I = ?

Output vector dimensionality: O = 10

One hidden layer is sufficent; For I-H-O node topology:

Number of training equations Neq = Ntrn*O

Number of unknown weights Nw = (I+1)*H+(H+1)*O

For robust solution want Neq >> Nw or

H << (Neq-O)/(I+O+1)

size(p) = [I Ntrn]

size(t) = [O Ntrn]

See NNTB documentation for examples.

For additional examples, search the Newsgroup using

heath newff close clear

Hope this helps.

Greg

Related Solutions

MATLAB: How to fix minmum mse requirement for neural network training mannually rather using default requirement

 [ I N ] = size(x)
 [ O N ] = size(t)

% Reference MSEs (From Naïve Constant Output Model y00 = mean(ttrn,2))

 Ntrn      = N-2*(0.15*N)       % Default no. of training examples
 MSEtrn00  = mean(var(ttrn',1)) % Biased training target variance (divide by Ntrn)
 MSEtrn00a = mean(var(ttrn',0)  % Degree of freedom "A"djusted (divide by Ntrn-1)

% Number of estimation degrees of freedom (See Wikipedia)

 Ntrneq    = Ntrn*O             % prod(size(ttrn)) is number of training equations
 Nw        = (I+1)*H+(H+1)*O    % Number of unknown weights for an I-H-O node topology
 Ndof      = Ntrneq-Nw          % Not useful when Nw >= Ntrneq
 etrn    = ttrn-ytrn;           % Training error
 MSEtrn  = sse(etrn)/Ntrneq     % Biased MSE
         = mse(etrn)
 MSEtrna = sse(etrn)/Ndof       % DOFA  MSE
         = Ntrneq*MSEtrn/Ndof

% Normalized MSEs

 NMSEtrn  = MSEtrn/MSEtrn00
 NMSEtrna = MSEtrna/MSEtrn00a

% Practical training goal

 NMSEtrna <= 0.01  % Independent of target scaling
 net.trainParam.goal = 0.01*Ndof*MSEtrn00a/Ntrneq

Hope this helps.

Thank you for formally accepting my answer

Greg

MATLAB: Normalized Frequency in Analog Filter Design.

Normalized frequency is for digital filters. If you want to design analog filters, they don't apply. Like you mentioned, you should use the syntax with 's' at the end. To get Wn, you can use buttord function (also has an 's' syntax for analog filter).

http://www.mathworks.com/help/signal/ref/buttord.html

Best Answer

Related Solutions

MATLAB: How to fix minmum mse requirement for neural network training mannually rather using default requirement

MATLAB: Normalized Frequency in Analog Filter Design.

Related Question