CODEWITHSUNDEEP
matlabneural-network
Neuronaut
Neuronaut

Reputation: 15

Matlab neural network toolbox save regression plot to file

I'm trying to run a matlab script (generated by nftool, as my matlab knowledge is poor at best) with a rather large data set through my ssh connection on my school's multi-core compute server. Since I can't directly look at the graphical interface that's produce while training an the network, I'd like to save the plots to a file (the one I think I want the most is the regression plot) so I can look at it after the job runs. I've only edited the code to automatically import the data files

% Solve an Input-Output Fitting problem with a Neural Network
% Script generated by NFTOOL
% Created Tue Nov 11 21:20:40 CST 2014
%
% This script assumes these variables are defined:
%
%   NNinput - input data.
%   NNoutput - target data.

% sets the same seed every time, so the rand() sequence is always identical
RandStream.setGlobalStream(RandStream('mt19937ar','seed',1));

close all % closes all of the figures that you have generated in your program
clear all % deletes all stored variables in your workspace
clc       % removes all lines in your command window

NNinput = load('NNinput');
NNoutput = load('NNoutput');

inputs = NNinput;
targets = NNoutput;

inputs = inputs.';
targets = targets.';

% Create a Fitting Network
hiddenLayerSize = 10;
net = fitnet(hiddenLayerSize);

% Choose Input and Output Pre/Post-Processing Functions
% For a list of all processing functions type: help nnprocess
net.inputs{1}.processFcns = {'removeconstantrows','mapminmax'};
net.outputs{2}.processFcns = {'removeconstantrows','mapminmax'};

% Setup Division of Data for Training, Validation, Testing
% For a list of all data division functions type: help nndivide
net.divideFcn = 'dividerand';  % Divide data randomly
net.divideMode = 'sample';  % Divide up every sample
net.divideParam.trainRatio = 70/100;
net.divideParam.valRatio = 15/100;
net.divideParam.testRatio = 15/100;

% For help on training function 'trainlm' type: help trainlm
% For a list of all training functions type: help nntrain
net.trainFcn = 'trainlm';  % Levenberg-Marquardt

% Choose a Performance Function
% For a list of all performance functions type: help nnperformance
net.performFcn = 'mse';  % Mean squared error

% Choose Plot Functions
% For a list of all plot functions type: help nnplot
net.plotFcns = {'plotperform','plottrainstate','ploterrhist', ...
  'plotregression', 'plotfit'};


% Train the Network
[net,tr] = train(net,inputs,targets);

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

% Recalculate Training, Validation and Test Performance
trainTargets = targets .* tr.trainMask{1};
valTargets = targets  .* tr.valMask{1};
testTargets = targets  .* tr.testMask{1};
trainPerformance = perform(net,trainTargets,outputs)
valPerformance = perform(net,valTargets,outputs)
testPerformance = perform(net,testTargets,outputs)

So far, all I've been able to come up with is...

h = findobj('Type', plotregression(targets, outputs), TRAINING_PLOTREGRESSION, 'regressionPlot');
for k = 1:numel(h)
    print(h(k), sprintf('Pic%d.ps',k));
end;

from this post how to save matlab neural networks toolbox generated figures

and I'm guessing that I would add this to the end of the file, but I'm pretty sure that that isn't right. If anybody can help me out it would be much appreciated!

Upvotes: 0

Views: 1874

Answers (1)

Pablo EM
Pablo EM

Reputation: 6679

This is likely the most basic way to save the plots about your training. You have chosen your plot functions in this part of your code:

% Choose Plot Functions
% For a list of all plot functions type: help nnplot
net.plotFcns = {'plotperform','plottrainstate','ploterrhist', ...
  'plotregression', 'plotfit'};

So now you can simply call each plot function at the end of your code (or anywhere after training the network) and save the plot with print:

plotperform(tr);
print('-dpsc', 'perform')
plottrainstate(tr);
print('-dpsc', 'trainstate')
ploterrhist(tr);
print('-dpsc', 'errhist')
plotregression(tr);
print('-dpsc', 'regression')
plotfit(tr);
print('-dpsc', 'fit')

The first argument of print selects the printer driver, in this case PostScript Level 3 color, and the second is the name of the figure. See here for more information about print.

Upvotes: 1

Related Questions