The following code removes duplicate values
numarr = [1 2 2 3 3 3 4 4 4 4];
%// Filter doubled values
[n_, bin] = histc(numarr, unique(numarr));
multiple_ = find(n_ > 1);
mult_indices_ = ismember(bin, multiple_);
numarr(mult_indices_) = [];
%// output
numarr = 1
How to adapt it, that the first occurence of any duplicate remains?
i.e. that the output will be
numarr =
1 2 3 4
Use unique with the 'stable' property:
a = [1 3 2 5 2 7 1 3 4 5 6 8 2];
output = unique(a,'stable')
which will keep the order and therefore keeps the first occurrance of every value, as desired.
output =
1 3 2 5 7 4 6 8
Without 'stable' the output gets sorted.
Regarding your comment: To get the indices of the removed duplicates you need the second output of unique and setdiff:
[output, kept_idx] = unique(a,'stable')
removed_idx = setdiff(1:numel(a),kept_idx)
removed_idx =
5 7 8 10 13
uniquefunction.