Now that we have independence idx &gnd why not direct comparison and use missclassGroups function generating permutations to compare,

Still want to join the ari in addition, the calculation of nmi, figuring out what the acc is the first step
Everyone know to share ~

 

Function [miss1, index]=missclassGroups (Segmentation, RefSegmentation ngroups) 
If (size (RefSegmentation, 2)==1) 
RefSegmentation=RefSegmentation '; 
End 
If (size (Segmentation, 2)==1) 
Segmentation=Segmentation '; % all become row vector 
End 
Permutations=perms (1: ngroups); 
Miss=zeros (size (Permutations, 1), size (Segmentation, 1)); 
For k=1: size (Segmentation, 1) % is actually only a layer of loop k=1 
For j=1: size (Permutations, 1) 
Miss (j, k)=sum (abs (Segmentation (k, :) - Permutations (j, RefSegmentation)) & gt; 0.1); 

% Permutations (j, RefSegmentation) like RefSeg dimension is equal to the number of samples 
% for the first j a full arrangement, according to the RefSeg (real value), the order of the order to withdraw from the Permutations (try Permutations (1,,5,0 [1])) 
% why ACC?? 
% is presupposed take absolute value & gt; 0.1 to judge independence idx and this arrangement out 640 predicted (a line of comments on whether the same, different is 1 
% sum for the second line comments is not the same as the number of, the higher the worse 

End 
End