Neural Network Input Size and Data Issue - HELP

Hi,

I am trying to code a feed forward neural network to predict future significant wave heights given known previous wave height values.

I have two inputs to my network. These are the wave heights over a year for two weather buoys. One input for my network is the wave heights over a year for one weather buoy (8744 data values), and the other input is the wave height values over one year for the second buoy (again 8744 data values).

I am coming up with the error 'Number of inputs does not match net.numInputs.' Can someone help spot my mistake please?

I am not sure what value I should be using in 'net.inputs{1}.size'.

My code is as follows:

%Data Selection
X = xlsread('2014_comp.xlsx', 'F5:F8748');
Y = xlsread('2014_comp.xlsx', 'L5:L8748');

%Network Architecture
net = feedforwardnet;                           % create feed forward network
net.numInputs = 2;                              % set number of inputs

net.inputs{1}.size = 1;                         % set size of input 1
net.inputs{2}.size = 1;                         % set size of input 2

net.numLayers = 1;                              % add 1 layer to network
net.layers{1}.size = 10;                        % assign number of neurons in layer
net.inputConnect(1) = 1;                        % connect input 1 to layer 1
net.inputConnect(2) = 1;                        % connect input 2 to layer 1

net.biasConnect(1) = 1;                         % connect bias to layer 1

net.biases{1}.learnFcn = 'learnp';              % set bias learning function
net.biases{1}.initFcn = 'initzero';             % set bias init function

net.outputConnect(1) = 1;                       %connect layer 1 to output

net.layers{1}.transferFcn = 'tansig';           % set layer transfer function to hyperbolic tangent sigmoid
net.inputWeights{1}.initFcn = 'initzero';       % set input wieghts init function
net.inputWeights{1}.learnFcn = 'learnp';        % set input weight learning function

net.initFcn = 'initlay';                        % set network init function
net.trainFcn = 'trainrp';                       % set network training function
net.performFcn = 'mse';                         % set network performance evaluation function

%Training/Testing/Validation Ratio
net.divideParam.trainRatio = 0.66;              % set proportion of data for training
net.divideParam.valRatio = 0.17;                % set proportion of data for validation
net.divideParam.testRatio = 0.17;               % set proportion of data for testing

view(net)

%Network Training
net = train(net,X);                           %Train the network

Thanks, James

Answer 1

[Greg Heath]

You are making this harder than it is. For the 1st time through all you have to specify is

 1. Type of network
 a fitnet for regression on curve-fitting
 b. patternnet for classification or pattern-recognition
 c. timedelaynet, narnet or narxnet for timeseries prediction
(fitnet and patternnet are special cases of feedforwardnet which never 
 has to be called explicitly)
 2. input % size(input)   = [ I N } for N I-dimensional Inputs
 3. target % size(target) = [ O N ] for N O-dimensional Output targets
 4. Ntrials: The number of candidate designs with different random data
 divisions and random initial weights.

Then modify the code in the help and/or doc documentation. For example, use the commands

help fitnet
and 
doc fitnet

to obtain code for fitnet. Put this code in a loop for ntrial = 1: Ntrials for different random data divisions and random initial weights.

This is the simplest code that I would use:

 [ x t ] = simplefit_dataset;
 vart    = var(t,1)% Reference MSE
 net     = fitnet;
 rng('default') % Needed for design duplication
 for i = 1:Ntrials
    net = configure(net,x,t);
    [ net tr y e ] = train(net,x,t);
    NMSE(i,1) = mse(e)/ vart;  % Normalized MSE
 end
 output = NMSE

All 10 designs will be terrific because this is a simple data set. For other data I usually try to find the minimum number of hidden nodes (H, fitnet(H)) that will yield NMSE <= 0.01 ( i.e., the net successfully models 99% of the target variance)

Hope this helps.

Thank you for formally accepting my answer

Greg

[James Helliwell]

Thank you! This is really useful!