• overview
• what's new
• download
• CORTEX stuff
• tutorial
• reference
• testing
• bugs and quirks

m@cortex testing

In order to ensure consistency of behavior between m@cortex updates, I will be mantaining a script that runs a number of checks on sample data files.

All interested users are encouraged to send me their own sample data files accompanied by short scripts explaining the type of check to run. I will embed in the master routine all user-contributed scripts, and will check new versions of the package against them all before posting it for download.

Thus, rather than taking my word for it, users will know that the package they are about to download has passed a number of tests, including their own.

Here is a simple example of a script you might want included for m@cortex testing:

c = cortex ('sample.0', 1:2);
k = key ('sample.key');
for i=1:k.len
   m(i) = meanfiringrate (c, 1, k(i));
end
l = load ('sample.mat');
if any (m ~= l.m)
   warning ('mean firing rate mismatch detected!);
end

In this particular case, along with the script, you would send me a raw CORTEX data file (sample.0), a text file to use as source for a @key object (sample.key), and a .mat file (sample.mat) containing a matrix (m) with a list of mean firing rates that you know to be correct.

Of course, there is no limit to the length and complexity of test scripts...