Thank you for the comment.. I just concern It should be same answer for 'feat' at Approximate coefficient.. From the example you given, YES the answer are same.. when I go through the code (getmswtfeat.m) It is look like different in how to determine the last answer.. for code (getmswpfeatV00.m).. I can understand the flow.. but for the 'getmswtfeat.m', I'm confused for term 'percentENER(notZER,:) = ...', '[cfs,longs] = wdec2cl(dec,'all')',.. and I hope the final answer for 'tab_ENER' should be same as WP code... sorry to many question...
I tried the code (getmswtfeat.m (Your WT code)) and try to remove and maintain 'log1p'.. the result for 'energy' approximate coefficient are different when using the code (getmswpfeatV00.m)..
If I run this code (getmswpfeatV00.m) and I only took the energy at approximation part (Low pass filter part).. It is the result will be same if I'm using this code (getmswtfeat.m (Your WT code))?
One more thing, actually the code below is for what? Because you said can comment it.. If I comment this code the result is different.
I test for J=1 and comment the code below and I compare with using this code ([CA,CD] = dwt(xx,'sym8');) and took the energy for CA and CD.. and the result is same.
%% You can comment this section
JJ = J;
for i=1:J-1
T = (2^i):(2^i)+(2^i)-1;
feat(:,T) = feat(:,T)./log(JJ);
JJ=JJ-1;
end
Hi. In this package, I did not include different features selection techniques. I will submit another package in the next month that includes about 4-5 feature selection methods. Meanwhile, some of them are already available in MatlabFileExchange (not by me! )
Comment only
27 Feb 2013
Ensemble Toolbox
This toolbox provides some combination methods to fuse an ensemble of classifiers
if you type: wilcoxon it will works showing the example.
in the other cases you have to write:
x1=[n1 n2 n3 n4 n5...ni];
x2=[m1 m2 m3 m4 m5...mi];
wilcoxon(x1,x2)
where n1...ni are the values of the first vector and m2...mi are the values of the second vector.
Comment only