how to avoid repeat numbers in a row
Mostra commenti meno recenti
Dear All,
I have this row
A=[1 2 3 4 5 6];
I tried to find random permutation on the vector but in many cases I got a row with repetition numbers for example
z=[1 5 2 4 1 5];or z=[1 5 2 4 1 1];
I want to avoid the repetition in this code but it is totally wrong, at first I should assign the missing numbers in such way, here is [3 6].
zc=[3 6 ];
e=size(zc,2);
z=[1 5 2 4 1 5];
for i=1:6
z(i)==z
s = find((z(i)==z) == 1)
ss=size(s,2);
for j=2:ss
for k=1:e
z(j)=zc(k)
end
end
end
I don't want to use a function from matlab to get the permutation, I need to fix my row or any permutation exchange file like (perm file ). If anyone help me for it and save me I will be grateful.
regards,
Imola
Risposta accettata
Più risposte (1)
Guillaume
il 19 Mar 2015
It seems to me that your effort would be better spent on fixing the code that generate these invalid permutation rather than fixing the permutation afterward.
Anyway, one way to do what you want:
z = [1 5 2 4 1 1];
zc = [3 6];
%build a cell array containing the indices of repeated values:
repindices = accumarray(z', 1:numel(z), [], @(v) {v(2:end)'})';
%convert cell array into vector:
repindices = [repindices{:}];
%replace values at indices by replacement:
z(repindices) = zc
4 Commenti
Guillaume
il 19 Mar 2015
Note that if you don't randomise which of the duplicates (my code above always pick the first ones) or the order of the replacements, then the distribution of the generated permutations is not uniform
imola
il 19 Mar 2015
Guillaume
il 19 Mar 2015
If you're asking how to build zc:
zc = setdiff(1:6, z)
imola
il 19 Mar 2015
Categorie
Scopri di più su Creating and Concatenating Matrices in Centro assistenza e File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!