custom multi layer feed forward neural network

4 次查看（过去 30 天）

sulaiman ahmed 2015-9-12

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/242575-custom-multi-layer-feed-forward-neural-network

评论： Greg Heath 2015-9-13

_ _ *I have custom multi layer feed forward program , but the validation and test performance equal to NaN value (ther is no curve , but only training performance ), i want simple code of using custom network for any type of inputs and outputs but i want introduce the validation and test performance value , I would be grateful if anyone can help me , the code is as follows :

close all, clear all, clc, format compact 
inputs = [1:6]' % input vector (6-dimensional pattern) 
outputs = [1 2]' % corresponding target output vector 
% create network 
net = network( ...
1, ... % numInputs, number of inputs,
2, ... % numLayers, number of layers
[1; 0], ... % biasConnect, numLayers-by-1 Boolean vector,
[1; 0], ... % inputConnect, numLayers-by-numInputs Boolean matrix,
[0 0; 1 0], ... % layerConnect, numLayers-by-numLayers Boolean matrix
[0 1] ... % outputConnect, 1-by-numLayers Boolean vector
);
net.layers{1}.size = 5; 
% hidden layer transfer function 
net.layers{1}.transferFcn = 'radbas';
view(net); 
net = configure(net,inputs,outputs); 
view(net); 
% initial network response without training 
initial_output = net(inputs)
% network training 
net.trainFcn = 'trainlm'; 
net.performFcn = 'mse'; 
[net,tr] = train(net,inputs,outputs);
% network response after training final_output = net(inputs)
plotperf(tr);