Rank: 13117 based on 0 downloads (last 30 days) and 0 file submitted
photo

Jan Stolarek

E-mail
Company/University
Technical University of Lodz

Personal Profile:

 

Watch this Author's files

 

Comments and Ratings by Jan Stolarek View all
Updated File Comments Rating
15 Dec 2011 MATLAB xUnit Test Framework MATLAB xUnit is a unit test framework for MATLAB code. Author: Steve Eddins

Steve >> There is no name pattern. I just throw my tests into one directory (which perhaps is not the best idea...) and I know that some of theme belong to one program, the others belong to another.

04 Dec 2011 MATLAB xUnit Test Framework MATLAB xUnit is a unit test framework for MATLAB code. Author: Steve Eddins

Hm... but how could I add many classes to a TestSuite using this?

02 Dec 2011 MATLAB xUnit Test Framework MATLAB xUnit is a unit test framework for MATLAB code. Author: Steve Eddins

I have a usage problem. I have some test classes that extend TestCase class. I would like to group some of these classes into a suite that runs them. How do I do that? I don't know how to use add method from TestSuite class. I came up with something like this:

suite = TestSuite( 'NameOfTheSuite' );
suite.add( testClassNo1Test( 'testMethodNo1' ) );
suite.add( testClassNo1Test( 'testMethodNo2' ) );
suite.add( testClassNo2Test( 'testMethodNo1' ) );
suite.add( testClassNo2Test( 'testMethodNo2' ) );

So basically I have to manually specify the name of every test method which is very tedious task. I don't want to use TestCase.fromPwd() - I want to create test suite containing only selected test cases.

09 Oct 2011 EZW (Embedded Zerotree Wavelet) Matlab implementation of EZW (Embedded Zerotree Wavelet) Author: Kanchi

Good code, clearly written and well commented but not optimized. After two days of refactoring I've managed to speed it up 2 times.

07 Oct 2011 SPIHT SPIHT matlab code (without Arithmatic coding stage) Author: Kanchi

Very nice code, although it could use some refactoring, e.g. this part in the main demo file:

OrigSize = size(Orig_I, 1);
max_bits = floor(rate * OrigSize^2);
OutSize = OrigSize;
image_spiht = zeros(size(Orig_I));
[nRow, nColumn] = size(Orig_I);
n = size(Orig_I,1);
n_log = log2(n);
level = n_log;

Could be simple rewritten as:

[nRow, nColumn] = size(Orig_I);
max_bits = floor(rate * nRow^2);
level = log2(nRow);

OutSize and image_spiht are never used in fact. Last line assumes silently that dimensions of the image are power of 2 and that there are at least as many columns as rows. Perhaps this would be better:
level = floor( log2( min( [ nRow, nColumn ] ) ) );

Matlab code analyzer complains a lot about variables growing inside the loops. I think that optimizing that could improve performance a lot.

Contact us