MATLAB: Input to a custom network

input to custom networkneural networkstraining of the custom network

I am designing a neural network with 4 inputs and 2 layers including the output layer. the input database is in an excel document and is of the matrix which has 97rows and 5 columns. This is in one sheet and I have 11 such sheets. I am able to import the data to the M-file; however I am unable to give it as an input to the neural network to train. How am I supposed to give my database as the input? Also using trainlm, it gives me an error saying zero input delay. How am I supposed to use the train function [net tr] if my and the input database is all those 11 sheets and the target is just one column?

Best Answer

% Read in the 11 sheets and combine to obtain a data matrix with size(data) = [ 5 1067 ] (1067 = 11*97)

close all, clear all, clc

input = data(1:4,:);

target = data(5,:);

[ I N ] = size(input) % [ 4 1067 ]

[ O N ] = size(target) % [ 1 1067 ]

% I = 4

% O = 1

% N = 1067

% input = randn( I ,N );

% hidden= sum(input);

% target = 1 + hidden+ hidden.^2 ;

% plot(hidden,target,'o')

% Using the NN default data division

Ntrn = N - 2*round(0.15*N) % 747

Neq = Ntrn*O % 747 Number of training equations

% Assuming an I-H-O = 4-H-1 FFMLP for regression or curvefitting, the number of unknown weights to be estimated by the 747 equations is

% Nw = (I+1)*H+(H+1)*O = O+(I+O+1)*H = 1+6*H.

% Typically, Neq > Nw is required although Neq >> Nw is desired. The resulting upper bound on H is

Hub = floor((Neq-O)/(I+O+1)) % 124

% Therefore, the default value H = 10 << Hub = 124 is a reasonable first choice.

net = fitnet;

net.trainParam.goal = var(target)/100;

[net tr Y E ] = train(net,input,target);

Hope this helps.

Greg

Related Solutions

MATLAB: How to calculate the error in Neural Network Back Propagation Network

Use as many defaults as possible

 MSEgoal = 0.01*mean(var(target',1))
 MinGrad = MSEgoal/100
 net = newff(input,target);
 net.trainParam.goal     = MSEgoal;
 net.trainParam.min_grad = MinGrad;
 [ net tr output error ] = train(net, input, target);

For details, search ANSWERS using the search word msegoal

Hope this helps.

Greg

Thank you for formally accepting my answer

Greg

MATLAB: Does anyone know why this neural network is not giving the right target values????plz any helpppppp

% Weird data set:

 plot(input,target,'o')

The data does not represent points from reasonable function. The target says" Regardless of the input, if the output is 0, your error will only be 3.33%".

Looking at an escalation of models (H=0:9) shows that the Linear model (H=0, NN with no hidden layer) is the best with a coefficient of determination (Wikipedia/R-squared) of only 0.0264. It is only slightly better than the constant output = 0 model.

 close all, clear all, clc
 N =30
 input  = [0 0 1 0 0 0 0 1 1 0 0 1 0 1 0 0 1 0 1 0 0 1 1 1 1 1 0 0 0 1]; 
 target = [ 1, zeros( 1,N-1)];

% The simplest model is the constant output model that outputs the mean of the target regardless of the input.

 y00   = (1/N)*ones(1,N); % 0.0333*ones(1,30)
 e00   = target-y00;     ; % [ 0.9667, -0.0333*ones(1,29)]
 MSE00 = mse(target-y00)       % 0.0322

% The next simplest is the linear model

 W0    =  target/[ ones(1,N) ; input ] % W0 = [ 0.0588 -0.0588 ]
 y0    =  W0 * [ ones(1,N) ; input ];  % y0 =  0.0588*(1-input)
 e0    =  target-y0;     ;             % [ 0.9412, -y0(2:end)]
 MSE0  =  mse(target-y0)               % 0.0314  Not much better than constant model
 R20   = 1 - MSE0/MSE00                %  0.0264   VERY BAD FIT

% Neural Model

 [ I N ] = size(input)                                 % [ 1 30 ]

 [ O N ] = size(target)                              % [ 1 30 ]
 Neq = N*O                                             % 30 =  No. of equations
% Nw = (I+1)*H+(H+1)*O = 1+3*H       % No. of unknown weights

% No overfitting Neq > Nw==> H <= Hub (upper bound)

 Hub  = -1+ceil( (Neq-O) / ( I+O+1) )  % 9
 Hmax = 9
 dH   = 1
 Hmin = 0  % Linear Model
 numH = length(Hmin:dH:Hmax) % 10
 Ntrials = 10
 rng(0)
 j=0
 for h=Hmin:dH:Hmax
    j=j+1
    if h==0
        net = patternnet([]);
        Nw = (I+1)*O
    else
        net = patternnet(h);
        Nw = (I+1)*h+(h+1)*O
    end
    for i=1:Ntrials
       net = configure(net, input,target);
       net.divideFcn = 'dividetrain';            % Default 'dividerand' is much worse
       [net tr y e ] = train(net,input,target);
       R2(i,j) = 1 - mse(e)/MSE00;
    end
 end
 R2 = R2   % 0.0264* ones(Ntrials,numH)     Same as Linear

Hope this helps.

*Thank you for formally accepting my answer *

Greg

Best Answer

Related Solutions

MATLAB: How to calculate the error in Neural Network Back Propagation Network

MATLAB: Does anyone know why this neural network is not giving the right target values????plz any helpppppp

Related Question