matlab read input command

I am working on NN and trying to code for 2:4:1 structure, please help me out in read input command as i am not able to give that command.

Thanks in Advance.

[KSSV]

Show us what you have tried.....

[PALI SAHU]

loadfile='samplefile.txt';
net=network;
net.numInputs= 2;
net.numLayers= 2;
net.layers{1}.size= 4;
net.layers{2}.size= 1;
net.inputconnect(1,2)= 2;
net.IW{1,1};
net.LW{2,1};
net.b{1};
net.b{2};
net.layerConnect = [0 0; 1 0]; 
net.biasConnect = [1;1]; 
net.outputConnect = [0 1];  
% Setup Division of Data for Training, Validation, Testing
%net.divideParam.trainRatio = 70/100;
%net.divideParam.valRatio = 15/100;
%net.divideParam.testRatio = 15/100;
net.divideFcn = 'dividerand';
% Define transfer Function for each layer
net.layers{1}.transferFcn = 'purelin';
net.layers{2}.transferFcn = 'logsig';
 % train the network
net.trainFcn= 'trainlm';

Answer 1

[Krishna]

Hi Pali,

I understand you want to code a neural network structure with 2 inputs, 1 hidden layer with 4 neurons, and 1 output using shallow networks. You can use the feedforwardnet function for this. Here's an example:

net = feedforwardnet([4], 'trainlm');

In this code, 'trainlm' is the optimizer, and 4 is the number of neurons in the hidden layer.

Next, format your dataset to have 2 inputs and 1 output. Then, use the ‘train’ function to train the neural network according to the required specifications.

Please go through the following documentation to learn more,

https://in.mathworks.com/help/deeplearning/ref/feedforwardnet.html

Also, apart from this MATLAB has dlnetworks which provide more flexibility on neural network design you can also try that.

Please look at this documentation to learn more about dlnetworks,

https://in.mathworks.com/help/deeplearning/ref/dlnetwork.html

Hope this helps.