Unit Tests

[kennethmorton]

The current unit test framework is jankety. The new MATLAB unit test framework introduced in 2013a seems like a good idea. Should we consider making the switch?

http://blogs.mathworks.com/steve/2013/03/12/matlab-software-testing-tools-old-and-new-r2013a/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+SteveOnImageProcessing+%28Steve+on+Image+Processing%29

[peterTorrione]

These unit test objects appear to be awesome. For example, the following will handle all our tests for classifiers, if we set it up properly. We can also make methods to automatically generate, say, the MAT files used to make sure the performance is good- e.g., "run this 20 times, and get me reasonable bounds, then save them".

The M-file is also in ]internal]unitTests, in the newest version, and should work with 2013A

classdef prtUnitTestClassifier < matlab.unittest.TestCase % prtUnitTestClassifier % Example test for classifiers. % % tester = prtUnitTestClassifier('classifier',prtClassFld,'perfLimsKfoldsUnimodal',[.9 1]) % fldResults = tester.run; % % tester = prtUnitTestClassifier('classifier',prtClassKnn,'perfLimsKfoldsUnimodal',[.95 1]) % knnResults = tester.run; % properties classifier perfLimsKfoldsUnimodal = [.9 1]; end

methods
    function self = prtUnitTestClassifier(varargin)
        self = prtUtilAssignStringValuePairs(self,varargin{:});
    end

end
methods (Test)
    function testKfoldsPercentCorrectUnimodal(self)

        ds = prtDataGenUnimodal;
        yOut = kfolds(self.classifier,ds,3);
        yOut = rt(prtDecisionBinaryMinPe,yOut);
        answer = prtScorePercentCorrect(yOut);
        self.verifyGreaterThan(answer,self.perfLimsKfoldsUnimodal(1));
        self.verifyLessThan(answer,self.perfLimsKfoldsUnimodal(2));

    end

    function testPlot2D(self)

        c = self.classifier.train(prtDataGenUnimodal);
        plot(c);

    end

    function testPlot3D(self)

        ds = catFeatures(prtDataGenUnimodal,prtDataGenUnimodal);
        ds = ds.retainFeatures(1:3);

        c = self.classifier.train(ds);
        plot(c);
    end
end

end