% Solve a Pattern Recognition Problem with a Neural Network
% Script generated by NPRTOOL
% Created Fri Jan 12 15:36:11 CET 2018
%
% This script assumes these variables are defined:
%
%   real_input - input data.
%   real_output - target data.

%load data
%load('out_in_pattern_classifier_2018_03_26.mat')

inputs = real_input;
targets = real_output;

% Create a Pattern Recognition Network
%hiddenLayerSize1 = 274;
hiddenLayerSize1 = 260;
%hiddenLayerSize2 = 165;
hiddenLayerSize2 = 156;
%net = patternnet(hiddenLayerSize1);
net = patternnet([hiddenLayerSize1 hiddenLayerSize2]);

%choose network algorithm
%net.trainFcn = 'trainscg';
net.trainFcn = 'trainrp';

% Setup Division of Data for Training, Validation, Testing
net.divideParam.trainRatio = 70/100;
net.divideParam.valRatio = 15/100;
net.divideParam.testRatio = 15/100;
%setup validation times:
%net.trainParam.max_fail = 100;
%net.trainParam.max_fail = 150;
net.trainParam.max_fail = 200;

% Train the Network
[net,tr] = train(net,inputs,targets); %redo this to get the full plot options with gui

% Test the Network
outputs = net(inputs);
errors = gsubtract(targets,outputs);
performance = perform(net,targets,outputs)

% View the Network
view(net)

% Plots
% Uncomment these lines to enable various plots.
%figure, plotperform(tr)
%figure, plottrainstate(tr)
%figure, plotconfusion(targets,outputs)
%figure, ploterrhist(errors)

%Choose save name
%save net1_new.mat
%save net2_new.mat
%save net3_new.mat
%save net4_new.mat
%save net5_new.mat
save net6_new.mat